La Condición Física como Determinante de Salud en Personas ... · Facultad de Medicina Universidad de Granada La Condición Física como Determinante de Salud en Personas Jóvenes

Departamento de Fisiología

Facultad de Medicina

Universidad de Granada

La Condición Física como Determinante de Salud en

Personas Jóvenes

Fitness as a Health Determinant in Young People


Jonatan Ruiz Ruiz

2007

A mis padres y hermana

Prof. Dr. Manuel J. CASTILLO GARZON Catedrático de Universidad --- Departamento de Fisiología FACULTAD DE MEDICINA Universidad de Granada

MANUEL J CASTILLO GARZÓN, CATEDRÁTICO DE FISIOLOGÍA

MÉDICA DE LA FACULTAD DE MEDICINA DE LA UNIVERSIDAD DE

GRANADA

CERTIFICA:

Que la Tesis Doctoral titulada: “La Condición Física como Determinante de

Salud en Personas Jóvenes” que presenta D. JONATAN RUIZ RUIZ al

superior juicio del Tribunal que designe la Universidad de Granada, ha sido

realizada bajo mi dirección durante los años 2002 a 2007, siendo expresión

de la capacidad técnica e interpretativa de su autor en condiciones tan

aventajadas que le hacen merecedor del Título de Doctor, siempre y cuando

así lo considere el citado Tribunal.

Fdo. Manuel J Castillo Garzón

Granada, 11 de Enero de 2007

Prof. Dr. Ángel Gutiérrez Sáinz

Profesor Titular de Universidad --- Departamento de Fisiología FACULTAD DE MEDICINA Universidad de Granada

ANGEL GUTIÉRREZ SÁINZ, PROFESOR TITULAR DE

UNIVERSIDAD DE LA FACULTAD DE MEDICINA DE LA

UNIVERSIDAD DE GRANADA

CERTIFICA:








Fdo. Ángel Gutiérrez Sáinz


MARÍA MARCELA GONZÁLEZ GROSS, PROFESORA TITULAR DE

UNIVERSIDAD DE LA FACULTAD DE CIENCIAS DE LA ACTIVIDAD

FÍSICA Y DEL DEPORTE DE LA UNIVERSIDAD POLITÉCNICA DE

MADRID

CERTIFICA:








Fdo. Mª Marcela González Gross


Prof Dr Marcela González-Gross VICEDECANA DE CALIDAD

Y ASUNTOS INTERNACIONALES

Departamento de Fisiología

Facultad de Medicina


La Condición Física como Determinante de Salud en

Personas Jóvenes

Fitness as a Health Determinant in Young People

Jonatan Ruiz Ruiz

Directores de Tesis

Miembros del Tribunal

Dr. Jose Viña Ribes Catedrático de Universidad Universidad de Valencia MD, PhD

Dr. Jose A López Calbet Profesor Titular de Universidad Universidad de Las Palmas MD, PhD

Dr. Alejandro Lucía Mulas Profesor de Universidad Universidad Europea de Madrid MD, PhD

Dr. Angelo Pietrobelli Profesor de Universidad Universidad de Verona MD, PhD

Dr. Carmen Adamuz Ruiz Directora del C.A.M.D. Junta de Andalucía MD, PhD

Granada, 22 de Febrero de 2007

Dr. Manuel J Castillo Garzón Catedrático de Universidad Universidad de Granada MD, PhD

Dr. Ángel Gutiérrez Sáinz Profesor Titular de Universidad Universidad de Granada MD, PhD

Dr. Marcela González Gross Profesor Titular de Universidad Universidad Politécnica de Madrid PhD

13

Contenidos [Contents]

Proyectos de Investigación [Research Projects] .............................................15

Resumen..........................................................................................................16

Summary ..........................................................................................................17

Abreviaturas [Abreviations] ..............................................................................18

Introducción [Introduction]................................................................................21

Objetivos ..........................................................................................................24

Aims..................................................................................................................25

Bibliografía [References]..................................................................................26

Material y Métodos [Material and Methods] ....................................................28

Resultados y Discusión [Results and Discussion] ..........................................31

I. Serum lipid and lipoprotein reference values of Spanish adolescents; The AVENA study. Ruiz JR, Ortega FB, Moreno LA, Warnberg J, Gonzalez-Gross M, Cano M, Gutierrez A, Castillo MJ, and the AVENA Study Group. Soz Praventiv Med 2006; 51: 99-109.

II. Serum lipids, body mass index and waist circumference during pubertal development in Spanish adolescents: The AVENA Study. Ruiz JR, Ortega FB, Tresaco B, Wärnberg J, Mesa JL, Gonzalez-Gross M, Moreno LA, Marcos A, Gutierrez A, Castillo MJ. Horm Metab Res 2006; 38: 832-837.

III. Health-related physical fitness assessment in childhood and adolescence; A European approach based on the AVENA, EYHS and HELENA studies. Ruiz JR, Ortega FB, Gutierrez A, Sjöström M, Castillo MJ. J Public Health 2006; 14: 269-277.

IV. Cardiorespiratory fitness is associated with features of metabolic risk factors in children. Should cardiorespiratory fitness be assessed in a European health monitoring system? The European Youth Heart Study. Ruiz JR, Ortega FB, Meusel D, Harro M, Oja P, Sjöström M. J Public Health 2006; 14: 94-102.

V. Cardiovascular fitness is negatively associated with homocysteine levels in female adolescents. Ruiz JR, Sola R, Gonzalez-Gross M, Ortega FB, Vicente-Rodriguez G, Garcia-Fuentes M, Gutierrez A, Sjöström M, Pietrzik K, Castillo MJ. Arch Pediatr Adolesc Med2007; 161: 166-171.

VI. Inflammatory proteins are associated with muscle strength in adolescents; The AVENA Study. Ruiz JR, Ortega FB, Wärnberg J, Moreno LA, Carrero JJ, Gonzalez-Gross M, Marcos A, Gutierrez A, Sjöström M. Submitted.

14

VII. Use of artificial neural network-based equation for estimating VO2max in adolescents. Ruiz JR, Ramirez-Lechuga J, Ortega FB, Benitez JM, Arauzo-Azofra A, Sanchez C, Sjöström M, Castillo MJ, Gutierrez A, Zabala M, on behalf of the HELENA Study Group. Submitted.

VIII. Hand span influences optimal grip span in male and female teenagers.Ruiz JR, España-Romero V, Ortega FB, Sjöström M, Castillo MJ, Gutierrez A. J Hand Surgery [Am] 2006; 31: 1367-72.

IX. A Mediterranean diet is not enough for health: physical fitness is an important additional contributor to health for the adults of tomorrow. Castillo-Garzon MJ, Ruiz JR, Ortega FB, Gutierrez-Sainz A. World Rev Nutr Diet 2007; 97: 114-38.

Conclusiones ...................................................................................................183

Conclusions .....................................................................................................184

Curriculum Vitae abreviado [Short CV] ...........................................................185

Agradecimientos [Acknowledgements] ...........................................................191

15

Proyectos de Investigación

El presente trabajo de investigación ha sido posible gracias a las subvenciones obtenidas en los siguientes proyectos de investigación:

Beca de Formación de Profesorado Universitario del Ministerio de Educación, Cultura y Deporte (AP2003-2128). Departamento de Fisiología de la Facultad de Medicina de la Universidad de Granada.

Estudio AVENA (Alimentación y Valoración del Estado Nutricional de los Adolescentes Españoles). Proyecto Nacional multicéntrico financiado por el Instituto de Salud Carlos III con Fondos de Investigación Sanitaria, Ministerio de Sanidad y Consumo (nº 00/0015), por el Consejo Superior de Deportes (05/UPB32/0, 109/UPB31/03 y 13/UPB20/04), por el Ministerio de Educación (AP2002-2920, AP2003-2128 y AP-2004-2745), y por varias empresas privadas: Panrico S.A., Madaus S.A., y Procter and Gamble S.A.

Estudio B12: Detección precoz de la deficiencia de vitamina B12 en población de riesgo. Proyecto financiado por el Instituto de Salud Carlos III con Fondos de Investigación Sanitaria (FIS PI021830).

Estudio EYHS (European Youth Heart Study). Proyecto Europeo realizado en Dinamarca, Estonia, Noruega, Portugal y Suecia. El doctorando sólo ha trabajado con los datos correspondientes a Estonia y Suecia. El estudio realizado en Estonia recibió financiación de Estonian Science Foundation No. 3277 and 5209, y el Estonian Centre of Behavioural and Health Sciences. El estudio realizado en Suecia recibió financiación por el Stockholm County Council.

Estudio HELENA (Healthy Lifestyle in Europe by Nutrition in Adolescence). Estudio financiado por la Comunidad Europea: European Community Sixth RTD Framework Programme (Contract FOOD-CT-2005-007034).

Ayudas a Grupos de Investigación de la Junta de Andalucía.

16

Resumen

Conocer la relación entre capacidad aeróbica, fuerza muscular y factores de riesgo de enfermedad cardiovascular en niños y adolescentes es de interés científico y sanitario. Además, para poder interpretar de una manera más precisa estas asociaciones es necesario disponer de una metodología sencilla y fiable. Esto puede ayudar a crear estrategias de prevención primaria desde las edades más tempranas.

El objetivo general de esta memoria de Tesis Doctoral es estudiar la relación entre condición física (especialmente capacidad aeróbica y fuerza muscular) y factores de riesgo de enfermedad cardiovascular en jóvenes, así como desarrollar nuevos métodos de estimación de la capacidad aeróbica y fuerza muscular en adolescentes.

Un total de 873 niños de 9 a 10 años y 971 adolescentes de 12 a 19 años conforman las poblaciones que han participado en los tres estudios de cohortes incluidos en la presente memoria de Tesis: El estudio AVENA (Alimentación y Valoración del Estado Nutricional de los Adolescentes Españoles), el EYHS (European Youth Heart Study), y el estudio HELENA (Healthy Lifestyle in Europe by Nutrition in Adolescence).

Los principales resultados de la memoria de Tesis sugieren que: a) La condición física se relaciona con parámetros de salud en niños y adolescentes. b) La capacidad aeróbica se asocia inversamente con factores tradicionales de enfermedad cardiovascular en niños de 9 a 10 años. c) La capacidad aeróbica se asocia con un factor novel de enfermedad cardiovascular tal como la homocisteína en niñas adolescentes, y esto tras ajustar por distintas variables de confusión incluido el genotipo MTHFR 677C>T. d) La fuerza muscular se asocia a proteínas de inflamación aguda tales como la proteína C reactiva en adolescentes. e) Se ha desarrollado y validado una nueva fórmula de estimación del consumo máximo de oxígeno a partir del resultado obtenido en el test de ida y vuelta de 20 metros, el sexo, la edad, el peso y la talla del adolescente. f) Hay un tamaño de agarre óptimo que debería ser ajustado en el dinamómetro cuando se evalúe la fuerza de prensión manual en adolescentes.

Los resultados de la presente memoria de Tesis muestran que la condición física en general y la capacidad aeróbica y la fuerza muscular en particular constituyen un importante marcador de salud en jóvenes, al igual que ya se había mostrado en adultos. Estos datos confirman la necesidad de incluir este tipo de mediciones en los sistemas educativos y de salud pública. El desarrollo de nuevos métodos de evaluación de la condición física para ser aplicados en estudios epidemiológicos ayudará a mejorar la calidad y el rigor de los mismos.

17

Summary

For public health strategies and preventive purposes, it is of interest to understand the associations between cardiorespiratory fitness, muscle strength and cardiovascular disease risk factors in children and adolescents. Development of new and more accurate methodology to assess cardiorespiratory fitness and muscle strength may help to better elucidate the links between fitness and health from on early ages.

The overall objective of this thesis was to examine the association between physical fitness (focused on cardiorespiratory fitness and muscle strength) and both traditional and novel cardiovascular disease risk factors in young populations, and to develop new methods to better estimate cardiorespiratory fitness and muscle strength in adolescents.

A total of 873 children (aged 9 to 10 years), and 971 adolescents (aged 12 to 19 years) from three different studies were involved in the present work: the AVENA study (Alimentación y Valoración del Estado Nutricional de los Adolescentes Españoles), the EYHS (European Youth Heart Study), and The HELENA (Healthy Lifestyle in Europe by Nutrition in Adolescence) Study.

The main outcomes were: a) Physical fitness is associated to a myriad of health parameters in young people. b) Cardiorespiratory fitness is inversely associated with traditional cardiovascular disease risk factors in children. c) Cardiorespiratory fitness is inversely associated with a novel cardiovascular disease risk factor, such as homocysteine levels in female adolescents after controlling for potential confounders including the MTHFR 677C>T genotype. d) Muscle strength is inversely associated with inflammatory proteins, such as C-reactive protein, in adolescents. e) A new equation to estimate maximum oxygen consumption from 20m shuttle run test performance (last half stage completed), sex, age, weight and height in adolescents has been developed and cross-validated. f) There is an optimal grip span to which the dynamometer should be adjusted when measuring handgrip strength in adolescents.

The results show that physical fitness, and especially cardiorespiratory fitness and muscle strength is an important health marker in also young people, as has been shown in adults. Health information systems should include monitoring of cardiorespiratory and muscle fitness among young individuals. Development of efficient methodology for large-scale collection of the cardiorespiratory and muscle fitness data may help to improve the quality and accuracy of the outcome.

18

Abreviaturas

ACSM American College of Sports Medicine

ANCOVA Analysis of covariance

ANN Artificial neural network

ANOVA Analysis of variance

Apo Apolipoprotein

AVENA Alimentación y Valoración del Estado Nutricional de los Adolescentes

BF Body fat

BMI Body mass index

BP Blood pressure

CD Compact disc

CRF Cardiorespiratory fitness

CVF Cardiovascular fitness

DNA Desoxirribonucleic acid

EYHS European Youth Heart Study

HDLc High density lipoprotein cholesterol

HOMA Homeostasis model assessment

HELENA Healthy Lifestyle in Europe by Nutrition in Adolescence

LDLc Low density lipoprotein cholesterol

Ln Natural logarithm

Lp(a) Lipoprotein (a)

MET Metabolic equivalent

MSE Mean sum of squared errors

MTHFR Methylenetetrahydrofolate reductase

PA Physical activity

RMSE Root mean sum of squared errors

RNA Ribonucleic acid

SD Standard deviation

SEE Standard error of estimate

SPSS Statistical Package for Social Sciences

SRT Shuttle run test

SSE Sum of squared errors

TC Total cholesterol

TG Triglycerides

VO2 Oxygen uptake

19

VO2max Maximum oxygen uptake

WC Waist circumference

W/H Waist to hip ratio

21

Introducción

En España, al igual que en el resto de los países occidentales, las enfermedades cardiovasculares constituyen la principal causa de muerte. Hay numerosos factores de riesgo para desarrollar enfermedad cardiovascular entre los que se incluyen una alteración del perfil lipídico, resistencia a la insulina, parámetros inflamatorios elevados, hipertensión, sobrepeso y obesidad. Tradicionalmente, la prevención y tratamiento de estos factores ha estado enfocada a la población adulta. No obstante, investigaciones recientes han puesto de manifiesto su incidencia en niños y adolescentes (Bao et al., 1995; Katzmarzyk et al., 2001; Srinivasan et al., 2002; Raitakari et al., 2003; Moreno et al., 2005). De hecho, existen evidencias científicas que indican que el inicio de la enfermedad cardiovascular se da en la adolescencia o incluso en la infancia, aún cuando las manifestaciones clínicas de la misma aparecen y alcanzan máxima relevancia en la vida adulta tardía (Berenson et al.,1998; Strong et al., 1999).

Son diversos los factores que pueden contribuir al inicio precoz, aunque subclínico, de la enfermedad cardiovascular. Entre esos factores se encuentran los cambios en los patrones alimenticios, descenso de los niveles de actividad física, aumento de los patrones de sedentarismo y otros. Estos patrones de comportamiento y su repercusión fisiológica se fijan principalmente durante la etapa adolescente. La adolescencia es una etapa decisiva en el desarrollo humano por los múltiples cambios fisiológicos y psicológicos que en ella ocurren. Este periodo se caracteriza por un intenso crecimiento, hasta el punto que se llega casi a duplicar el peso corporal del niño. A esto contribuye también el desarrollo sexual, el cual va a desencadenar importantes cambios en la composición corporal del niño. Por otro lado, se producen importantes cambios psicológicos que tienden a afectar su imagen corporal, la forma de alimentarse y el modo de comportarse. Con frecuencia, los hábitos que comienzan en la adolescencia (tales como fumar, consumir alcohol, comer de manera saludable o hacer ejercicio) suelen persistir durante muchos años o incluso durante toda la vida.

Estimaciones recientes sugieren que tanto la falta de actividad física como una dieta no saludable son dos claros factores de riesgo no sólo para desarrollar enfermedad cardiovascular sino para desarrollar muchas otras enfermedades. Ambos factores se cree que son responsables de alrededor de 400.000 muertes por año en Estados Unidos (Mokdad et al., 2004). Estas cifras están cerca de sobrepasar al tabaco como causa de mortalidad prevenible, y es previsible que la situación sea similar en el resto de los países occidentales.

Un factor íntimamente ligado al nivel de actividad física y/o ejercicio que se realiza es el estado de condición física que tiene la persona. La condición física se define como la capacidad que una persona tiene para realizar ejercicio. La condición física constituye una medida integrada de todas las funciones y estructuras que intervienen en la realización de activad física o ejercicio. Estas funciones son la músculo-esquelética, cardio-respiratoria, hemato-circulatoria, endocrino-metabólica y psico-neurológica. Un alto nivel de condición física implica una buena respuesta fisiológica de todas ellas. Por el contrario, tener una mala condición física podría indicar un malfuncionamiento de una o varias de esas funciones.

22

La condición física comprende un conjunto de cualidades físicas tales como la capacidad aeróbica, fuerza y resistencia muscular, movilidad articular, velocidad de desplazamiento, agilidad, coordinación, equilibrio, y composición corporal. La medición de estas cualidades físicas en estudios epidemiológicos es relativamente reciente, y su aplicación al ámbito de la salud ha originado el sobrenombre de condición física relacionada con la salud (en inglés health-related fitness).

Condición física y salud

De todas las cualidades que componen la condición física, la capacidad aeróbica, la fuerza muscular y la composición corporal han sido las que han adquirido una mayor relevancia científica en el ámbito sanitario. No obstante, la relación del resto de cualidades físicas con distintos parámetros de salud también ha sido reconocida en personas jóvenes y adultas (American College of Sports Medicine, 1998).

Condición física y salud: capacidad aeróbica

La capacidad aeróbica (en inglés aerobic capacity, cardiorespiratory fitness, cardiovascular fitness) es una de las cualidades más importantes de la condición física relacionada con la salud. La capacidad aeróbica representa una medida directa del estado general de salud y de manera específica del estado del sistema cardiovascular, respiratorio y metabólico.

Recientes investigaciones han puesto de manifiesto el interés que tiene conocer el nivel de capacidad aeróbica que posee una persona. Tener un nivel medio-alto de capacidad aeróbica disminuye el riesgo de desarrollar enfermedad cardiovascular y aumenta la esperanza de vida en adultos (Blair et al., 1989; Lee et al., 1999; Carnethon et al., 2005; LaMonte et al., 2005). Asimismo, una mejora de la capacidad aeróbica se asocia directamente con una mejora de la calidad de vida no sólo en personas sanas sino también en personas con cáncer (Herrero et al., 2006). La capacidad aeróbica también se ha asociado inversamente con distintos parámetros de salud en jóvenes, tales como el perfil lipídico, la resistencia a la insulina, la masa grasa, parámetros relacionados con el síndrome metabólico y la resistencia arterial (Gonzalez-Gross et al., 2003; Gutin et al., 2004; Eisenmann et al., 2005; Gutin et al., 2005; Reed et al., 2005; Mesa et al., 2006; Ruiz et al., 2006).

Condición física y salud: fuerza muscular

El papel de la fuerza muscular en la práctica de ejercicio y actividades de la vida diaria, así como en la prevención de diversas enfermedades está siendo objeto de creciente atención en los último años (Stump et al., 2006; Wolfe, 2006). La fuerza muscular se puede mejorar mediante el entrenamiento contrarresistencia, ejercicio que está recomendado por importantes organizaciones relacionadas con la salud para mejorar la condición física y la salud tanto de personas sanas como de personas con alguna enfermedad (Pollock et al., 2000; Kraemer et al., 2002).

La fuerza muscular se ha asociado inversamente con distintos parámetros relacionados con el síndrome metabólico (i.e. triglicéridos, lipoproteínas de alta densidad, glucosa, tensión arterial, y circunferencia de cintura) en hombres (Jurcaet al., 2004), así como con proteínas de inflamación aguda en hombres y mujeres (Visser et al., 2002; Schaap et al., 2006).

23

Además, resultados de estudios prospectivos han mostrado que aquellos hombres que tenían mejor fuerza muscular tenían también menor incidencia de síndrome metabólico, y esto tras ajustar por varios parámetros de confusión entre los que se incluía la capacidad aeróbica (Jurca et al., 2005). Los resultados de un estudio longitudinal en los que se siguió durante 40 años a más de 1.000 hombres, mostraron que una baja fuerza de prensión manual se asociaba a un mayor índice de morbilidad y mortalidad por todas las causas independientemente del nivel de actividad física y de la masa muscular de los participantes (Metter et al., 2002). Estos resultados muestran la importancia de mantener unos niveles de fuerza muscular relativamente altos para mantener una buena calidad de vida y reducir la incidencia de morbilidad.

Por todo ello, es de capital importancia desarrollar herramientas de diagnóstico y prevención a aplicar ya desde edades tempranas para identificar alteraciones en aquellos factores que puedan incrementar el riesgo de desarrollar alguna enfermedad cardiovascular durante estos años y en la vida adulta.

Con base en estos antecedentes, la presente memoria de Tesis fija los siguientes objetivos:

24

Objetivos

General:

El objetivo general de la Tesis Doctoral es estudiar la relación entre condición física (especialmente capacidad aeróbica y fuerza muscular) y factores de riesgo de enfermedad cardiovascular en jóvenes, así como desarrollar nuevos métodos de estimación de la capacidad aeróbica y fuerza muscular en adolescentes.

Específicos:

I. Describir el estado de salud del adolescente español en lo que referente a los niveles de lípidos y lipoproteínas sanguíneas.

II. Describir la influencia de la edad cronológica y el desarrollo madurativo durante la adolescencia sobre los niveles de lípidos y lipoproteínas sanguíneas, el índice de masa corporal y la circunferencia de la cintura.

III. Analizar la relación existente entre condición física y estado de salud en jóvenes.

IV. Estudiar la asociación entre la capacidad aeróbica y los factores tradicionales de riesgo de enfermedad cardiovascular en niños de 9 a 10 años.

V. Estudiar la asociación entre la capacidad aeróbica y un factor novel de riesgo de enfermedad cardiovascular, la homocisteína.

VI. Estudiar la relación entre la fuerza muscular y parámetros de inflamación, examinando si esta asociación se ve influenciada por el peso corporal en adolescentes.

VII. Desarrollar una ecuación basada en el modelo de redes neuronales para mejorar la estimación de la capacidad aeróbica en estudios poblacionales en adolescentes.

VIII. Determinar si el tamaño de la mano de los adolescentes influye sobre la media de la fuerza de prensión manual.

IX. Discutir la relación entre dieta, actividad física, condición física y parámetros de riesgo cardiovascular en niños y adolescentes.

25

Aims

Overall:

The overall objective of this thesis was to examine in young populations the association between physical fitness (focused on cardiorespiratory fitness and muscle strength) and both traditional and novel cardiovascular disease risk factors, and to develop new methods to better estimate cardiorespiratory fitness and muscle strength in adolescents.

Specific:

I. To provide current reference values for serum lipid and lipoprotein levels in Spanish adolescents according to age and sex.

II. To describe the effects of chronological age and pubertal development on serum lipid and lipoprotein levels, body mass index and waist circumference in Spanish adolescents.

III. To study the association between physical fitness and health in young people.

IV. To examine the associations between cardiorespiratory fitness and metabolic risk factors in children aged 9 to 10 years.

V. To examine the association between cardiorespiratory fitness and homocysteine levels in adolescents.

VI. To analyse the associations between inflammatory proteins and muscle strength, and to determine whether these associations vary in overweight and non-overweight adolescents.

VII. To develop an artificial neural network-based equation for estimating maximum oxygen consumption in adolescents.

VIII. To determine if there is an optimal grip span for determining the maximum handgrip strength in adolescents.

IX. To study the associations between physical activity, fitness and cardiovascular disease risk factors from on early ages.

26

Bibliografía

American College of Sports Medicine (1998) American College of Sports Medicine Position Stand. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Med Sci Sports Exerc 30, 975-991.

Bao W, Threefoot SA, Srinivasan SR & Berenson GS (1995) Essential hypertension predicted by tracking of elevated blood pressure from childhood to adulthood: the Bogalusa Heart Study. Am J Hypertens 8, 657-665.

Berenson GS, Srinivasan SR, Bao W, Newman WP, 3rd, Tracy RE & Wattigney WA (1998) Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study. N Engl J Med 338, 1650-1656.

Blair SN, Kohl HW, 3rd, Paffenbarger RS, Jr., Clark DG, Cooper KH & Gibbons LW (1989) Physical fitness and all-cause mortality. A prospective study of healthy men and women. Jama 262, 2395-2401.

Carnethon MR, Gulati M & Greenland P (2005) Prevalence and cardiovascular disease correlates of low cardiorespiratory fitness in adolescents and adults. Jama 294, 2981-2988.

Eisenmann JC, Katzmarzyk PT, Perusse L, Tremblay A, Despres JP & Bouchard C (2005) Aerobic fitness, body mass index, and CVD risk factors among adolescents: the Quebec family study. Int J Obes (Lond) 29, 1077-1083.

Gonzalez-Gross M, Ruiz JR, Moreno LA, De Rufino-Rivas P, Garaulet M, Mesana MI, Gutierrez A & Group A (2003) Body composition and physical performance of Spanish adolescents: the AVENA pilot study. Acta Diabetologica 40 Suppl 1, S299-301.

Gutin B, Yin Z, Humphries MC, Bassali R, Le NA, Daniels S & Barbeau P (2005) Relations of body fatness and cardiovascular fitness to lipid profile in black and white adolescents. Pediatr Res 58, 78-82.

Gutin B, Yin Z, Humphries MC, Hoffman WH, Gower B & Barbeau P (2004) Relations of fatness and fitness to fasting insulin in black and white adolescents. J Pediatr 145, 737-743.

Herrero F, Balmer J, San Juan AF, Foster C, Fleck SJ, Perez M, Canete S, Earnest CP & Lucia A (2006) Is cardiorespiratory fitness related to quality of life in survivors of breast cancer? J Strength Cond Res 20, 535-540.

Jurca R, Lamonte MJ, Barlow CE, Kampert JB, Church TS & Blair SN (2005) Association of muscular strength with incidence of metabolic syndrome in men. Med Sci Sports Exerc 37, 1849-1855.

Jurca R, Lamonte MJ, Church TS, Earnest CP, Fitzgerald SJ, Barlow CE, Jordan AN, Kampert JB & Blair SN (2004) Associations of muscle strength and fitness with metabolic syndrome in men. Med Sci Sports Exerc 36, 1301-1307.

Katzmarzyk PT, Perusse L, Malina RM, Bergeron J, Despres JP & Bouchard C (2001) Stability of indicators of the metabolic syndrome from childhood and adolescence to young adulthood: the Quebec Family Study. J Clin Epidemiol 54, 190-195.

Kraemer WJ, Adams K, Cafarelli E, Dudley GA, Dooly C, Feigenbaum MS, Fleck SJ, Franklin B, Fry AC, Hoffman JR, Newton RU, Potteiger J, Stone MH, Ratamess NA & Triplett-McBride T (2002) American College of Sports Medicine position stand. Progression models in resistance training for healthy adults. Med Sci Sports Exerc34, 364-380.

LaMonte MJ, Barlow CE, Jurca R, Kampert JB, Church TS & Blair SN (2005) Cardiorespiratory fitness is inversely associated with the incidence of metabolic syndrome: a prospective study of men and women. Circulation 112, 505-512.

Lee CD, Blair SN & Jackson AS (1999) Cardiorespiratory fitness, body composition, and all-cause and cardiovascular disease mortality in men. American Journal of Clinical Nutrition 69, 373-380.

Mesa JL, Ruiz JR, Ortega FB, Warnberg J, Gonzalez-Lamuno D, Moreno LA, Gutierrez A & Castillo MJ (2006) Aerobic physical fitness in relation to blood lipids and fasting glycaemia in adolescents: Influence of weight status. Nutr Metab Cardiovasc Dis 16, 285-293.

27

Metter EJ, Talbot LA, Schrager M & Conwit R (2002) Skeletal muscle strength as a predictor of all-cause mortality in healthy men. J Gerontol A Biol Sci Med Sci 57, B359-365.

Mokdad AH, Marks JS, Stroup DF & Gerberding JL (2004) Actual causes of death in the United States, 2000. Jama 291, 1238-1245.

Moreno LA, Mesana MI, Fleta J, Ruiz JR, Gonzalez-Gross M, Sarria A, Marcos A, Bueno M & Group AS (2005) Overweight, obesity and body fat composition in spanish adolescents. The AVENA Study. Annals of Nutrition & Metabolism 49, 71-76.

Pollock ML, Franklin BA, Balady GJ, Chaitman BL, Fleg JL, Fletcher B, Limacher M, Pina IL, Stein RA, Williams M & Bazzarre T (2000) AHA Science Advisory. Resistance exercise in individuals with and without cardiovascular disease: benefits, rationale, safety, and prescription: An advisory from the Committee on Exercise, Rehabilitation, and Prevention, Council on Clinical Cardiology, American Heart Association; Position paper endorsed by the American College of Sports Medicine. Circulation 101, 828-833.

Raitakari OT, Juonala M, Kahonen M, Taittonen L, Laitinen T, Maki-Torkko N, Jarvisalo MJ, Uhari M, Jokinen E, Ronnemaa T, Akerblom HK & Viikari JS (2003) Cardiovascular risk factors in childhood and carotid artery intima-media thickness in adulthood: the Cardiovascular Risk in Young Finns Study. Jama 290, 2277-2283.

Reed KE, Warburton DE, Lewanczuk RZ, Haykowsky MJ, Scott JM, Whitney CL, McGavock JM & McKay HA (2005) Arterial compliance in young children: the role of aerobic fitness. Eur J Cardiovasc Prev Rehabil 12, 492-497.

Ruiz JR, Rizzo NS, Hurtig-Wennlof A, Ortega FB, Warnberg J & Sjostrom M (2006) Relations of total physical activity and intensity to fitness and fatness in children: the European Youth Heart Study. Am J Clin Nutr 84, 299-303.

Schaap LA, Pluijm SM, Deeg DJ & Visser M (2006) Inflammatory markers and loss of muscle mass (sarcopenia) and strength. Am J Med 119, 526 e529-517.

Srinivasan SR, Myers L & Berenson GS (2002) Predictability of childhood adiposity and insulin for developing insulin resistance syndrome (syndrome X) in young adulthood: the Bogalusa Heart Study. Diabetes 51, 204-209.

Strong JP, Malcom GT, McMahan CA, Tracy RE, Newman WP, 3rd, Herderick EE & Cornhill JF (1999) Prevalence and extent of atherosclerosis in adolescents and young adults: implications for prevention from the Pathobiological Determinants of Atherosclerosis in Youth Study. Jama 281, 727-735.

Stump CS, Henriksen EJ, Wei Y & Sowers JR (2006) The metabolic syndrome: role of skeletal muscle metabolism. Ann Med 38, 389-402.

Visser M, Pahor M, Taaffe DR, Goodpaster BH, Simonsick EM, Newman AB, Nevitt M & Harris TB (2002) Relationship of interleukin-6 and tumor necrosis factor-alpha with muscle mass and muscle strength in elderly men and women: the Health ABC Study. J Gerontol A Biol Sci Med Sci 57, M326-332.

Wolfe RR (2006) The underappreciated role of muscle in health and disease. Am J Clin Nutr84, 475-482.

28

Material y Métodos

El material y métodos de la presente memoria de Tesis se basan fundamentalmente en proyecto AVENA. Se presenta una copia del resumen del artículo metodológico de dicho proyecto. Además, se muestra una tabla resumen de la metodología utilizada en los artículos que componen la presente memoria de Tesis.

ResumenAntecedentes: La adolescencia es una etapa decisiva en el desarrollo humano por los múltiples cambios fisiológicos y psicológicos que en ella ocurren los cuales, a su vez, condicionan tanto las necesidades nutricionales como los hábitos de alimentación, actividad física y comportamiento. Además, está demostrado que estos hábitos tienen repercusión en el estado de salud en la vida adulta. El interés de este tema así como su apropiado desarrollo ha merecido una financiación por parte del Fondo de Investigación Sanitaria del Instituto de Salud Carlos III. Objetivo: Desarrollar una metodología que evalúe el estado de salud así como la situación nutricional-metabólica y forma física de una muestra representativa de adolescentes españoles. Especial atención se prestará a tres tipos específicos de patologías como son obesidad, anorexia nerviosa/bulimia, dislipidemia.Metodología: Para alcanzar el objetivo, se van a estudiar ocho tipos diferentes de magnitudes: 1) ingesta dietética, hábitos alimentarios y conocimientos nutricionales; 2) actividad física habitual y actitud frente a la práctica físico-deportiva; 3) nivel de condición física; 4) antropometría y composición corporal; 5) estudio hematobioquímico: perfil fenotípico lipídico y metabólico, estudio hematológico; 6) perfil fenotípico de factores lipídicos de riesgo cardiovascular; 7) perfil inmunológico de estado nutricional; 8) perfil psicológico.Conclusión: Este proyecto incluye la actividad coordinada de cinco centros españoles situados en otras tantas ciudades (Granada, Madrid, Murcia, Santander, Zaragoza). Cada uno de esos centros tiene larga y acreditada experiencia en la parte del estudio de la que es responsable. En función de los resultados obtenidos, se propondrá un programa específico de intervención que permita mejorar la alimentación y neutralizar el riesgo que, para las patologías antes mencionadas, existe entre los adolescentes españoles. Con ello se pretende contribuir a mejorar el estado de salud de la población española del nuevo milenio.

Abstract Background: Adolescence is a decisive period in human life due to the multiple physiological and psychological changes that take place. These changes will condition both nutricional requirements and eating/physical activity behavior. It has been demonstrated that these “adolescence” factors are of significant influence in health status during adult life. Due to its importance and adequate development the project has been granted by the Fondo de Investigación Sanitaria of the Institute of Health Carlos III. Objective: To develop a methodology to evaluate the health and nutritional status of a representative population of Spanish adolescents. Specific attention is paid to three specific health problems: obesity, anorexia nervosa/bulimia, dislipidemia. Methodology: The following magnitudes will be studied: 1) dietary intake, food habits and nutrition knowledge; 2) daily physical activity and personal approach; 3) physical condition; 4) anthropometry and body composition; 5) hematobiochemical study: plasma lipid phenotypic and metabolic profile, blood cell counts; 6) genotypic profile of cardiovascular risk lipid factors; 7) immune function profile related to nutritional status; 8) psychological profile. Conclusion: This project includes the co-ordinate activity of five Spanish centers of five different cities (Granada, Madrid, Murcia, Santander, Zaragoza). Each center is specialized in a specific area and will be responsible for the corresponding part of the study. From the data obtained, we will elaborate a specific intervention program in order to improve nutrition and neutralize the risk for nutritional related problems in adolescence. By this, we will contribute to improve the health status of the Spanish population in the new millennium.

29

Pro

yec

to

Art

ícu

lo

Dis

eño

S

uje

tos

Va

ria

ble

s es

tud

iad

as

Met

od

olo

gía

AV

EN

A

AV

EN

A

AV

EN

A

EY

HS

HE

LE

NA

EY

HS

AV

EN

A

B12

AV

EN

A

HE

LE

NA

HE

LE

NA

AV

EN

A

EY

HS

HE

LE

NA

I. S

erum

lip

id a

nd l

ipop

rote

in r

efer

ence

val

ues

of S

pani

sh

adol

esce

nts;

The

AV

EN

A s

tudy

II. S

erum

lip

ids,

bod

y m

ass

inde

x an

d w

aist

cir

cum

fere

nce

duri

ng

pube

rtal

dev

elop

men

t in

Spa

nish

ado

lesc

ents

: T

he A

VE

NA

Stu

dy

III.

Hea

lth-

rela

ted

phys

ical

fit

ness

ass

essm

ent

in c

hild

hood

and

ad

oles

cenc

e; A

Eur

opea

n ap

proa

ch b

ased

on

the

AV

EN

A, E

YH

S

and

HE

LE

NA

Stu

dies

IV. C

ardi

ores

pira

tory

fit

ness

is

asso

ciat

ed w

ith

feat

ures

of

met

abol

ic

risk

fac

tors

in

chil

dren

. Sho

uld

car

dior

espi

rato

ry f

itne

ss b

e as

sess

ed

in a

Eur

opea

n he

alth

mon

itor

ing

syst

em?

The

Eur

opea

n Y

outh

Hea

rt

Stu

dy

V. C

ardi

ovas

cula

r fi

tnes

s is

neg

ativ

ely

asso

ciat

ed w

ith

hom

ocys

tein

e le

vels

in

fem

ale

adol

esce

nts

VI.

Inf

lam

mat

ory

prot

eins

are

ass

ocia

ted

wit

h m

uscl

e st

reng

th i

n ad

oles

cent

s; T

he A

VE

NA

Stu

dy

VII

. Use

of

arti

fici

al n

eura

l ne

twor

k-ba

sed

equa

tion

for

est

imat

ing

VO

2max

in

adol

esce

nts

VII

I. H

and

span

inf

luen

ces

opti

mal

gri

p sp

an i

n m

ale

and

fem

ale

teen

ager

s

IX.A

Med

iter

rane

an d

iet

is n

ot e

noug

h fo

r he

alth

: ph

ysic

al f

itne

ss i

s an

im

port

ant

addi

tion

al c

ontr

ibut

or t

o he

alth

for

the

adu

lts

of

tom

orro

w

Tra

nsve

rsal

Tra

nsve

rsal

Rev

isió

n

Tra

nsve

rsal

Tra

nsve

rsal

Tra

nsve

rsal

Tra

nsve

rsal

Art

ícul

o m

etod

ológ

ico

Exp

erim

enta

l A

rtíc

ulo

met

odol

ógic

o

Rev

isió

n

299

niño

s

282

niña

sE

dad:

13-

18.5

254

niño

s

272

niña

s E

dad:

13-

18.5

Niñ

os y

ad

oles

cent

es

429

niño

s

444

niña

s E

dad:

9-1

0

76 n

iños

80

niñ

as

Eda

d: 1

3-18

.5

230

niño

s 18

6 ni

ñas

Eda

d: 1

3-18

.5

122

niño

s 71

niñ

as

Eda

d: 1

3-19

101

niño

s 10

6 ni

ñas

Eda

d: 1

3-18

Niñ

os y

ad

oles

cent

es

TG

, TC

, HD

Lc,

LD

Lc,

apo

A-1

, Apo

B-1

00,

Lp(

a) y

eda

d de

men

arqu

ia

TG

, TC

, HD

Lc,

LD

Lc,

TC

, apo

A-1

, apo

B-

100,

Lp(

a), e

dad

de m

enar

quia

, Tan

ner,

BM

I y

WC

CA

, fle

xibi

lida

d, f

uerz

a m

uscu

lar,

vel

ocid

ad d

e m

ovim

ient

o, a

gili

dad,

y v

ario

s pa

rám

etro

s de

sa

lud

TG

, TC

, HD

Lc,

HO

MA

, TA

, MG

y C

RF

CA

, MG

, tH

cy, M

TH

R 6

77C

>T

, fol

ato

y vi

tam

ina

B12

sér

ico,

Tan

ner,

pes

o al

nac

er,

SE

S y

cig

arri

llos

CR

P, C

3,C

4, c

erul

opla

smin

a y

tran

sthy

reti

na,

CA

, fue

rza

mus

cula

r, B

MI,

MG

, Tan

ner

y S

ES

CA

, pes

o y

tall

a

Fue

rza

de p

rens

ión

man

ual

y ta

mañ

o de

la

man

o

CA

, fue

rza

mus

cula

r, a

ctiv

idad

fís

ica,

die

ta y

pa

rám

etro

s de

sal

ud c

ardi

ovas

cula

r

Ana

liza

dor

enzi

mát

ico

está

ndar

e

inm

unon

efel

omet

ría

Ana

liza

dor

enzi

mát

ico

está

ndar

, in

mun

onef

elom

etrí

a, c

uest

iona

rios

, pes

o, t

alla

y

cint

a m

étri

ca

Rev

isió

n bi

blio

gráf

ica

y co

ntex

tual

izac

ión

de

resu

ltad

os p

ropi

os

Ana

liza

dor

enzi

mát

ico

está

ndar

, pli

egue

s cu

táne

os, t

ensi

ómet

ro a

utom

átic

o, t

est

máx

imo

en c

iclo

ergó

met

ro

20m

SR

T, p

lieg

ues

cutá

neos

, téc

nica

de

PC

R,

inm

unoe

nsay

o, y

cue

stio

nari

os

Aná

lisi

s es

tánd

ar p

or i

nmun

otur

bodi

omet

ría,

20

mS

RT

, din

amom

etrí

a m

anua

l, s

alto

de

long

itud

a p

ies

junt

os, p

eso,

tal

la y

cu

esti

onar

ios

20m

SR

T, m

edid

or d

e ga

ses

port

átil

Din

amóm

etro

man

ual

Rev

isió

n bi

blio

gráf

ica

y co

ntex

tual

izac

ión

de

resu

ltad

os p

ropi

os

TG

: tr

igli

céri

dos,

TC

: co

lest

erol

tot

al, H

DL

c: l

ipop

rote

ínas

de

alta

den

sida

d, L

DL

c: l

ipop

rote

ínas

de

alta

den

sida

d, A

po:

apol

ipop

rote

ínas

, Lp:

lip

opro

teín

a, H

OM

A:

hom

eost

asis

mod

el a

sses

smen

t, t

Hcy

: ho

moc

iste

ína,

TA

: te

nsió

n ar

teri

al, C

RP

: pr

oteí

na C

rea

ctiv

a, B

MI:

índ

ice

de m

asa

corp

oral

, SE

S:

esta

tus

soci

econ

ómic

o, M

G:

mas

a gr

asa,

CA

: ca

paci

dad

aeró

bica

, 20m

SR

T:

test

de

ida

y vu

elta

de

20 m

etro

s, P

CR

: po

lim

eras

e ch

aín

reac

tion

.

Tab

la 1

. Res

umen

de

la m

etod

olog

ía u

tili

zada

en

los

artí

culo

squ

e co

mpo

nen

la p

rese

nte

mem

oria

de

Tes

is.

31

Resultados y Discusión

Los resultados y discusión se presentan a continuación en la forma en que han sido previamente publicados/sometidos en revistas científicas.

REFERENCE VALUES FOR SERUM LIPIDS AND

LIPOPROTEIN IN SPANISH ADOLESCENTS

THE AVENA STUDY

Jonatan R. Ruiz1, Francisco B. Ortega1, Luis A Moreno2, Julia

Wärnberg3,4, Marcela Gonzalez-Gross1,5, Maria D. Cano6, Ángel

Gutiérrez1, Manuel J. Castillo1, and the AVENA Study Group

Soz Praventiv Med 2006; 51: 99-109

1Departamento de Fisiología, Facultad de Medicina, Universidad de Granada, Granada, Spain. 2E.U. Ciencias de la Salud, Universidad de Zaragoza, Zaragoza, Spain. 3Grupo Inmunonutrición, Departamento de Nutrición y

Metabolismo, Consejo Superior de Investigaciones Científicas, Madrid, Spain. 4Unit for Preventive Nutrition, Department of Biosciences at Novum, Karolinska

Institutet, Huddinge, Stockholm, Sweden. 5Facultad de CC. de la Actividad Física y el Deporte. Universidad Politécnica de Madrid, Madrid, Spain. 6Sección de Lípidos del Hospital Clínico Universitario, Granada, Spain.

I

Soz Praventiv Med. 51 (2006) 99–1090303-8408/06/020099–11DOI 10.1007/s00038-005-0021-9© Birkhäuser Verlag, Basel, 2006

Jonatan R Ruiz1, Francisco B Ortega1, Luis A Moreno2, Julia Wärnberg3,4, Section: International Comparison of Health DeterminantsMarcela Gonzalez-Gross1,5*, Maria D Cano6, Angel Gutierrez1,Manuel J Castillo1, and the AVENA Study Group

1 Departamento de Fisiología, Facultad de Medicina, Universidad de Granada, Granada, Spain 2 EU Ciencias de la Salud, Universidad de Zaragoza, Zaragoza, Spain3 Grupo Inmunonutrición, Departamento de Nutrición y Metabolismo, Consejo Superior de Investigaciones Científi cas, Madrid, Spain4 Unit for Preventive Nutrition, Department of Biosciences at Novum, Karolinska Institutet, Huddinge, Stockholm, Sweden5 Facultad de CC. de la Actividad Física y el Deporte. Universidad Politécnica de Madrid, Madrid, Spain6 Sección de Lípidos del Hospital Clínico Universitario, Granada, Spain* At the time of the study, MGC was with (3)

Reference values for serum lipids and lipoproteins in Spanish adolescents: the AVENA study

Submitted: May 18, 2005

Accepted: December 12, 2005

Summary

Objectives: To provide current reference values for serum lipid

and lipoprotein levels in Spanish adolescents according to age

and sex.

Methods: A cross sectional study conducted in five representa-

tive Spanish cities (Granada, Madrid, Murcia, Santander and

Zaragoza) including a representative sample of 581 adolescents

(299 male and 282 female), aged 13 to 18.5 years. Age- and

sex-specific means, standard deviations and percentiles were

determined for: Total (TC), high density lipoprotein (HDLc) and

low density lipoprotein (LDLc) cholesterol, triglycerides, apoli-

poprotein A-1 and B-100, and lipoprotein(a).

Results: The 90th percentile for TC was 4.95 mmol/L for males

and 5.19 mmol/L for females. HDLc levels were significantly

higher in females of all age groups. LDLc levels ranged from

2.32 to 2.54 mmol/L in males and from 2.38 to 2.62 mmol/L in

females, peaking at 13 years of age in both sexes. Triglyceride

levels tended to increase gradually and to peak at 17 years

of age for both sexes. Apolipoprotein A-1 and B-100 levels

paralleled those of HDLc and LDLc values, respectively. The

geometric mean for lipoprotein(a) levels ranged from 0.44 to

0.57 μmol/L in males and from 0.50 to 0.67 μmol/L in females.

Conclusions: The present study provides reference data on the

distribution of lipid and lipoprotein levels of Spanish adoles-

cents.

Keywords: Adolescents – Lipids – Lipoproteins – Cardiovascular disease – Percentiles

Coronary heart disease (CHD) is a leading cause of global mortality, accounting for almost 17 million deaths every year

(Smith et al. 2004). Nearly 80 % of this mortality and disease burden occurs in the industrialized countries; the data for Spain reflect this picture (Instituto Nacional de Estadistica, 2001). Pathological data have shown that atherosclerosis begins in childhood (Berenson et al. 1998; Strong et al. 1999), and CHD is known to occur more frequently in adult members of families in which children’s cholesterol levels are high. Aortic fatty streaks can be found in children, and fibrousplaques are often evident in adolescence (McGill et al. 1997). This finding, plus the alarming increase in the prevalence of obesity (Moreno et al. 2002; Moreno et al. 2005) and the re-duction in physical activity among children and adolescents (Kimm et al. 2002; Moreno et al. 2002; Tercedor 2003), shows the need for improved health education in this age group (Gaziano et al. 1998). The relationship between serum lipids and the development of CHD in children and adoles-cents is well established (Berenson et al. 1998).The meta-analysis performed by Plaza (1991) showed that the serum total cholesterol (TC) levels of Spanish children and adolescents increased throughout the 1980s. However, no current data on serum lipid or lipoprotein levels data are avail-able. The AVENA Study was therefore designed to asses the health and nutritional status of a representative population of Spanish adolescents. This report describes the current serum lipid and lipoprotein profiles of Spanish adolescents living in urban areas, and compares the results with those obtained in other countries.

Materials and methods

Population and sample recruitmentThe methodology used in this study has been described

100 Section: International Comparison of Health Determinants Ruiz JR, Ortega FB, Moreno LA, et al. Reference values for serum lipids and lipoproteins in Spanish

adolescents: the AVENA study

Soz Praventiv Med. 51 (2006) 99–109© Birkhäuser Verlag, Basel, 2006

elsewhere (Gonzalez-Gross et al. 2003a, b; Moreno et al. 2005). Briefly, a multicenter study was performed involving a representative sample of Spanish adolescents aged from 13 to 18.5 years. The population was selected by multiple-step, simple random sampling – first taking into account location (Madrid, Murcia, Granada, Santander and Zaragoza) and then by random assignment of the school within each city. The cit-ies were chosen according to the population rate (>100 000 inhabitants), geographical location in the country (north-south gradient, in order to be representative) and taking into account the main technical question, that is, the necessity of having a research group in the city. Sample size was stratifiedby age and sex. The socio-economic variable was considered to be associated to location within the city and type of school. As the selection of schools was done by random selection proportionally to the number of schools in each city district, guaranteeing the presence of almost one school per district, the socio-economic variable was also considered to be ran-domly assigned. After analysis of the data, this method has proven to be adequate, as the socio-economic status of our sample has a normal distribution according to the distribution in the Spanish society.To calculate the number of adolescents to be included in the study in order to guarantee a representative sample of the whole country, we selected the variable with the greatest variance for this age group from the data published in the literature at the time the study was planned; that was body mass index (BMI) (Moreno et al. 1997). The sampling was determined for the distribution of this variable; the CI was established at 95 % with an error ±0.25 %. The minimum subject population was established at 1 750 for the complete study and at 500 for a subgroup from whose member’s blood samples were required. A similar number of subjects was evaluated in each city, and proportionally distributed by sex and age group (13, 14, 15, 16, 17–18.5 years).The sample was oversized in order to prevent loss of infor-mation and because technically it was necessary to perform fieldwork in complete classrooms. After finishing the field-work, the subjects who did not fulfill the inclusion criteria were excluded. Finally, the sample was adjusted by a weight factor in order to balance the sample in accordance to the distribution of the Spanish population and to guarantee the real representativeness of each of the groups, already definedby the previously mentioned factors (age and sex). The finalnumber of subjects included in the AVENA Study was 2 859 adolescents, from which 581 (299 males and 282 females) had blood measurements, and were then included in this study.In each school all the adolescents of one classroom were proposed to participate in the survey. A detailed verbal de-scription of the nature and purpose of the study was given

to both the children and their teachers. This information was also sent to parents by letter; written consent to be included was requested from both parents and children. The exclusion criteria were: no personal history of cardiovascular or meta-bolic disease; free of disease and medication at the time of the study; pregnancy. In order to avoid a selection bias, a family history record of metabolic and cardiovascular diseases was obtained for all subjects participating in the study.The protocol for the complete multicenter study was approved by the Review Committee for Research Involving Human Subjects of the Hospital Universitario Marqués de Valdecilla (Santander, Spain).

Blood measurementsBlood (20 ml) was collected from an antecubital vein between 8:00 and 9:00 a.m, after an overnight fast.

Measurement of serum lipids, lipoproteins and lipoprotein(a)Total cholesterol (TC), triglycerides (TG) and high density lipoprotein cholesterol (HDLc) were measured by enzymatic assay using a Hitachi 911 Analyzer (Roche Diagnostics, Indianapolis, Ind, USA). HDLc was precipitated before analysis using the Boehringer Mannheim method. Low den-sity lipoprotein cholesterol (LDLc) was calculated using the Friedewald et al. (1972) formula adjusted for serum TG levels (Morley et al. 1998). Apolipoprotein (apo) A-1, apo B-100 and lipoprotein(a) [Lp(a)] were measured by immunoneph-elometric assay using an Array 306 system (Beckman GMI, Inc., Albertville, Minnesota, USA). Quality control of the assays was assured by the Regional Health Authority. The coefficients of variation were less than 3 % and the intra-class coefficients were higher than 0.96 % for all blood variables. The following atherogenic indices were also calculated: TC/HDLc, TC-HDLc, (TC-HDLc)/HDLc, TG/HDLc, LDLc/HDLc, apo B-100/apo A-1, and apo B-100/LDLc. Age at menarche was determined from the self-reported date of first menses based on administered questionnaire.

Statistical analysisFor data analysis, the studied population was divided into five age groups: 13–13.99, 14–14.99, 15–15.99, 16–16.99 and 17–18.5 years. Age- and sex-specific means, standard deviations (SD) and percentiles were determined. Kol-mogorov-Smirnov test was used to check data distribution by both sex and age and only by sex. The studied variables were quasi-normal distributed, but the asymmetry and kur-tosis levels were adequate for all, except for Lp(a) that was achieved after logarithmic transformation. Mean values were compared with one way analysis of variance (ANOVA), and post hoc Bonferroni test. The Mann-Whitney U test was used

Ruiz JR, Ortega FB, Moreno LA, et al. Section: International Comparison of Health Determinants 101Reference values for serum lipids and lipoproteins in Spanish adolescents: the AVENA study


to determine any differences in BMI (the variable selected to calculate the number of subjects to be included in the study) between the subgroup from which blood samples were obtained (N = 581) and the remaining subjects (N = 2278) (for each age subgroup and sex). No differences were seen between any of the age and sex groups (Tab. 1). The error was fixed at 0.05.

Results

The means and SDs for lipid and lipoprotein levels, according to age and sex are shown in Table 2. The percentiles distribu-tions for lipid and lipoprotein levels and atherogenic indices, according to age and sex are shown in tables 3–10. Compari-

son between the sexes shows both higher TC and HDLc lev-els in females than in males adolescents. Higher LDLc levels were only observed in females aged 15 years (P < 0.05). The differences in apo A-1 and apo B-100 levels between the sexes were entirely superimposable on those for HDLc and LDLc levels. Triglycerides levels were slightly lower in females al-though the differences failed to reach statistical significance,except for the 14 year-olds.An 8.2 % decline in mean TC serum levels was observed in males between the ages of 13 and 15 years (P < 0.05). For males 13 years of age, the 90th percentile for TC (5.39 mmol/L) was the highest estimate for all age groups and both sexes. For females, mean serum TG levels were no different among age groups. For females aged 17–18.5 years, the 90th

Sex Age group (years) Body Mass Index P =

Blood group (N = 581) Non Blood group (N = 2278)

Male 13 20.6 ± 3.3 20.6 ± 4.0 0.63

14 22.1 ± 3.9 21.4 ± 3.6 0.19

15 22.3 ± 4.0 21.9 ± 3.5 0.74

16 21.7 ± 3.3 21.8 ± 3.1 0.86

17–18.5 23.6 ± 4.2 22.7 ± 3.5 0.15

Female 13 21.0 ± 3.9 21.7 ± 3.6 0.08

14 21.4 ± 4.2 21.2 ± 3.5 0.35

15 21.3 ± 3.2 21.5 ± 3.0 0.67

16 21.9 ± 3.2 21.6 ± 3.1 0.52

17–18.5 21.8 ± 2.9 21.7 ± 3.3 0.68

Table 1 Comparisons of body mass index between sub-group in which blood sample was obtained (blood group) and group in which blood sample was not obtained (non blood group). Body mass index was calculated as body weight (kg) without shoes and with light clothing, divided by height (m) squared.

Table 2 Lipids and lipoprotein mean and SD values in Spanish adolescents aged 13 to 18.5 years. Values are means ± SD. TC: total cholesterol; HDLc: high density lipoprotein cholesterol; LDLc: low density lipoprotein cholesterol; TG: triglycerides; Apo: apolipoprotein; Lp(a): lipoprotein a. Geometricmean ± SD. aP < 0.05 for differences between sexes. *P < 0.05 (in comparison to males 15 years of age). ¶P < 0.05 (in comparison to males 17 years of age). #P < 0.05 (in comparison to males 13, 14 and 15 years of age).

Age groups (years) TC (mmol/L) HDLc (mmol/L) LDLc (mmol/L) TG (mmol/L) Apo A-1 (g/L) Apo B (g/L) Lp(a)û (μmol/L)

Males

13 4.26 ± 0.80* 1.35 ± 0.29a 2.54 ± 0.66 0.82 ± 0.41 1.16 ± 0.17a 0.71 ± 0.18 0.44 ± 0.06

14 4.02 ± 0.59a 1.32 ± 0.27a 2.32 ± 0.54 0.84 ± 0.41a 1.10 ± 0.19 0.67 ± 0.15 0.49 ± 0.05

15 3.91 ± 0.60a 1.31 ± 0.23a¶ 2.24 ± 0.54a 0.78 ± 0.28 1.12 ± 0.20a 0.65 ± 0.14a 0.49 ± 0.04

16 4.07 ± 0.64 1.41 ± 0.27a 2.30 ± 0.59 0.79 ± 0.33 1.26 ± 0.20# 0.68 ± 0.13 0.48 ± 0.06

17–18.5 4.01 ± 0.73a 1.23 ± 0.18a 2.39 ± 0.71 0.86 ± 0.37 1.20 ± 0.17a 0.70 ± 0.16 0.57 ± 0.06

Total (13–18.5) 4.05 ± 0.68 1.32 ± 0.25a 2.35 ± 0.62 0.82 ± 0.36 1.17 ± 0.19 0.68 ± 0.15 0.49 ± 0.05

Females

13 4.51 ± 0.59 1.53 ± 0.27 2.62 ± 0.52 0.78 ± 0.24 1.24 ± 0.15 0.71 ± 0.11 0.59 ± 0.05

14 4.32 ± 0.70 1.53 ± 0.28 2.48 ± 0.67 0.69 ± 0.27 1.14 ± 0.26 0.72 ± 0.16 0.52 ± 0.05

15 4.38 ± 0.63 1.57 ± 0.33 2.48 ± 0.56 0.72 ± 0.23 1.28 ± 0.24 0.70 ± 0.13 0.55 ± 0.05

16 4.23 ± 0.69 1.53 ± 0.32 2.38 ± 0.58 0.69 ± 0.24 1.29 ± 0.23 0.68 ± 0.13 0.50 ± 0.05

17–18.5 4.40 ± 0.76 1.51 ± 0.28 2.51 ± 0.65 0.83 ± 0.64 1.34 ± 0.22 0.73 ± 0.15 0.67 ± 0.05

Total (13–18.5) 4.37 ± 0.68 1.53 ± 0.30 2.49 ± 0.60 0.74 ± 0.37 1.26 ± 0.23 0.71 ± 0.14 0.56 ± 0.05




percentile for TG (1.69 mmol/L) was the highest for all age groups and both sexes. For males, mean LDLc levels tended to decrease gradually from 13 to 15 years of age (11.6 %, P = 0.07). In males, the means and percentiles for serum apo A-1 showed distributions similar to those observed for HDLc. In females, apo A-1 levels increased significantly from 14 to 17–18.5 years of age. The percentile distributions for apolipo-protein B-100 were similar to those observed for LDLc, with no differences among age groups for either males or females. Differences were seen, however, between 15 year-old males and females (P < 0.05). No gender or age group differences

were found in Lp(a) levels. For the atherogenic indices, the TG/HDLc index was significantly higher in males than in females at age 14 and 15 years. Among females, the apo B-100/apo A-1 ratio was significantly higher at 14 compared to 15 years of age. The self-reported age of menarche ranged from 9 to 15 years of age. The age at first menses distribution was: 9 years (1.6 %), 10 years (2.9 %), 11 years (20.9 %), 12 years (34.7 %), 13 years (27.4 %), 14 years (11.3 %), and 15 years (1 %). No differences were observed in serum lipids variables within these groups.

Age groups (years) Total cholesterol

N Mean 10th 25th 50th 75th 90th

Males

13 54 4.26 3.32 3.70 4.20 4.69 5.39

14 54 4.02 3.32 3.65 3.94 4.46 4.86

15 63 3.91 3.29 3.56 3.86 4.19 4.68

16 63 4.07 3.34 3.60 4.03 4.43 4.80

17–18.5 65 4.01 3.06 3.46 4.11 4.51 4.95

Total (13–18.5) 299 4.05 3.29 3.57 3.99 4.49 4.95

Females

13 50 4.51 3.69 4.14 4.45 4.98 5.23

14 55 4.32 3.43 3.78 4.32 4.88 5.23

15 55 4.38 3.68 3.94 4.27 4.78 5.33

16 59 4.23 3.36 3.76 4.12 4.90 5.14

17–18.5 63 4.40 3.51 3.87 4.33 4.94 5.20

Total (13–18.5) 282 4.37 3.52 3.86 4.33 4.87 5.19

Table 3 Mean and percentile distributions for total choles -terol (mmol/L) according to ageand sex group. To convert cholesterol values in mmol/L to mg/dL divided by 0.02586.

Age groups (years) High density lipoprotein cholesterol


Males

13 54 1.35 0.93 1.11 1.40 1.50 1.68

14 54 1.32 0.91 1.14 1.32 1.53 1.65

15 63 1.31 1.04 1.17 1.32 1.47 1.60

16 63 1.41 1.03 1.24 1.40 1.58 1.79

17–18.5 65 1.23 1.04 1.06 1.22 1.37 1.48

Total (13–18.5) 299 1.32 1.02 1.14 1.32 1.49 1.66

Females

13 50 1.53 1.16 1.36 1.55 1.68 1.93

14 55 1.53 1.11 1.35 1.50 1.79 1.94

15 55 1.57 1.18 1.37 1.52 1.71 1.99

16 59 1.53 1.09 1.26 1.53 1.72 2.05

17–18.5 63 1.51 1.18 1.30 1.45 1.71 1.90

Total (13–18.5) 282 1.53 1.14 1.32 1.53 1.71 1.95

Table 4 Mean and percentile distributions of high density lipoprotein cholesterol (mmol/L) according to age and sex group. To convert cholesterol values in mmol/L to mg/dL divided by 0.02586.



Discussion

This study provides national reference data for the serum lipid and lipoprotein levels of Spanish adolescents living in urban areas. The percentile distributions according to age and sex are also established. To our knowledge, this is the first report to record the entire serum lipid and lipoprotein profile of a representative sample of Spanish adolescents ranging from 13 to 18 years. The mean TC, TG and LDLc levels of the present adoles-cents were similar or slightly lower than those observed in the meta-analysis of Plaza (1991), and then later by Garcés

et al. (2004). HDLc levels were also slightly lower than those observed two decades ago (Plaza 1991), perhaps due to a loss of Mediterranean dietary patterns (Moreno et al. 2002; Serra-Majen et al. 1995; Zamora et al. 2003) or to the low level of physical activity recorded for the Spanish population (Moreno et al. 2002) and adolescents of the AVENA study (Tercedor 2003). According to the NHANES III study, the mean serum TC lev-els of American children and adolescents aged 12–19 years were 4.09 mmol/L and 4.33 mmol/L for males and females respectively (Hickman et al. 1998). The present Spanish ado-

Table 5 Mean and percentile distributions of low density lipoprotein cholesterol (mmol/L) according to age and sex group. To convert cholesterol values in mmol/L to mg/dL divided by 0.02586.

Age groups (years) Low density lipoprotein cholesterol


Males

13 54 2.54 1.73 2.07 2.48 2.86 3.39

14 54 2.32 1.67 1.97 2.27 2.61 3.04

15 63 2.24 1.54 1.95 2.19 2.53 2.88

16 63 2.30 1.69 1.92 2.25 2.56 2.98

17–18.5 65 2.39 1.35 1.88 2.37 3.04 3.37

Total (13–18.5) 299 2.35 1.66 1.95 2.31 2.72 3.20

Females

13 50 2.62 1.87 2.32 2.59 2.94 3.30

14 55 2.48 1.64 1.96 2.45 2.97 3.33

15 55 2.48 1.79 2.09 2.46 2.82 3.27

16 59 2.38 1.67 2.02 2.36 2.72 3.35

17–18.5 63 2.51 1.60 2.02 2.51 2.98 3.27

Total (13–18.5) 282 2.49 1.74 2.07 2.46 2.92 3.30

Table 6 Mean and percentile distributions for apolipoprotein A-1 (g/L) according to age and sex group. To convert apo lipo-protein A-1 values in g/L to mg/dL divided by 0.01.

Age groups (years) Apolipoprotein A-1


Males

13 54 1.16 0.95 1.02 1.14 1.29 1.37

14 54 1.10 0.82 0.99 1.13 1.25 1.31

15 63 1.12 0.93 1.02 1.12 1.25 1.37

16 61 1.26 1.02 1.12 1.21 1.40 1.55

17–18.5 57 1.20 0.96 1.06 1.17 1.35 1.43

Total (13–18.5) 290 1.17 0.96 1.04 1.16 1.29 1.42

Females

13 50 1.24 1.04 1.15 1.24 1.36 1.43

14 55 1.14 0.64 1.06 1.21 1.29 1.42

15 55 1.28 1.02 1.13 1.27 1.46 1.58

16 48 1.29 1.02 1.16 1.27 1.44 1.61

17–18.5 58 1.34 1.10 1.19 1.30 1.45 1.71

Total (13–18.5) 267 1.26 1.02 1.13 1.25 1.38 1.56




lescents (both males and females) show TC levels similar to those of their American counterparts. The age and sex specifictrends for TC levels recorded in the present study were also similar to those reported in the NHANES III study (Hickman et al. 1998). Compared with data from Greece (Shulpis & Karikas 1998), another Mediterranean country, the Spanish mean serum TC levels were slightly higher. The NHANES III (Hickman et al. 1998) and LRC study (Kwiterovich 1991) re-ported higher TC levels in females than in males; this was also found in the present study. The NHANES III and LRC preva-lence studies showed lower TC levels among males during

puberty as a result of a decrease in HDLc levels (Hickman et al. 1998; Kwiterovich 1991). This agrees with that seen in the AVENA study. This reduction probably stems from hormonal changes experienced by males during puberty (Kwiterovich 1991). In the present adolescents, the HDLc levels were high-er than those recorded for American adolescents (Hickman et al. 1998). This might be attributable to genetic factors, en-vironmental factors and/or to the consumption of olive oil, a major component of the Mediterranean diet (Serra-Majem et al. 1993a, b; Moreno et al. 2002). Therefore, despite having a TC similar to that of American adolescents, the higher HDLc

Age groups (years) Apolipoprotein B-100


Males

13 54 0.71 0.49 0.60 0.70 0.79 0.96

14 54 0.67 0.46 0.56 0.64 0.75 0.91

15 63 0.65 0.49 0.57 0.64 0.72 0.87

16 61 0.68 0.51 0.57 0.68 0.75 0.84

17–18.5 57 0.70 0.49 0.60 0.72 0.82 0.91

Total (13–18.5) 290 0.68 0.49 0.58 0.67 0.77 0.88

Females

13 50 0.71 0.58 0.64 0.71 0.79 0.85

14 55 0.72 0.51 0.60 0.73 0.82 0.93

15 55 0.70 0.52 0.61 0.69 0.78 0.91

16 48 0.68 0.49 0.57 0.69 0.75 0.88

17–18.5 58 0.73 0.54 0.61 0.75 0.83 0.88

Total (13–18.5) 267 0.71 0.53 0.61 0.71 0.80 0.88

Table 7 Mean and percentile distributions for apolipoprotein B-100 (g/L) according to age and sex group. To convert apolipo-protein B values in g/L to mg/dL divided by 0.01.

Table 8 Mean and percentile distributions for triglycerides (mmol/L) according to age and sex group. To convert triglyce-rides values in mmol/L to mg/dL divided by 0.01125.

Age groups (years) Triglycerides


Males

13 54 0.82 0.36 0.49 0.73 1.11 1.41

14 54 0.84 0.46 0.59 0.70 1.02 1.43

15 63 0.78 0.47 0.57 0.71 0.90 1.16

16 63 0.79 0.42 0.55 0.77 0.94 1.23

17–18.5 65 0.86 0.52 0.64 0.77 1.00 1.36

Total (13–18.5) 299 0.82 0.44 0.58 0.75 0.96 1.31

Females

13 50 0.78 0.45 0.58 0.76 0.94 1.11

14 55 0.69 0.38 0.50 0.63 0.83 1.00

15 55 0.72 0.45 0.57 0.68 0.86 1.16

16 59 0.69 0.47 0.53 0.62 0.82 0.94

17–18.5 63 0.83 0.40 0.51 0.68 0.81 1.69

Total (13–18.5) 282 0.74 0.44 0.54 0.68 0.84 1.09



levels of the Spanish youngsters may renders them a healthier lipid profile. Their HDLc levels were, however, lower than those observed in Greek male adolescents aged 13 and 14 years (Shulpis & Karikas 1998). The HDLc levels recorded for females aged 13 and 14 years in the present study were the same as those of the Greek schoolchildren (Shulpis & Ka-rikas 1998). The HDLc levels of Spanish females were higher than that of males; which agree with the results reported by other authors (Azizi et al. 2001). HDLc is a protective factor for females; it is estimated that for every 0.0259 mmol/L (1

mg/dL) increase in HDLc, the risk of a CHD event is reduced by at least by 3 % in females, and 2 % in men (Nicklas et al. 1997).Low density lipoprotein cholesterol is the main carrier of cho-lesterol in the blood, and this compound plays a pivotal role in atherogenesis. The mean LDLc levels of Spanish adolescents were similar to those reported in American adolescents (Hick-man et al. 1998) but much higher than in Greek adolescents (Shulpis & Karikas 1998). In contrast, Spanish adolescents had much lower TG values than those observed in either

Table 9 Mean and percentile distributions for lipoprotein (a) (μmol/L) according to age and sex group. To convert lipo-protein (a) values in μmol/L to md/dL divided by 0.0357. Geometric mean.

Age groups (years) Lipoprotein (a)

N Meanû 10th 25th 50th 75th 90th

Males

13 54 0.44 0.04 0.17 0.54 1.46 3.19

14 54 0.49 0.04 0.21 0.54 1.59 2.83

15 61 0.49 0.07 0.21 0.44 1.02 2.55

16 51 0.48 0.05 0.14 0.40 1.70 3.58

17–18.5 63 0.57 0.10 0.27 0.55 2.45 3.25

Total (13–18.5) 284 0.49 0.05 0.21 0.48 1.50 3.00

Females

13 50 0.59 0.08 0.28 0.80 1.58 3.71

14 55 0.52 0.05 0.23 0.61 1.74 2.91

15 55 0.55 0.07 0.18 0.49 1.37 3.35

16 55 0.50 0.07 0.21 0.37 1.14 3.00

17–18.5 63 0.67 0.04 0.22 0.57 1.38 3.55

Total (13–18.5) 279 0.56 0.07 0.22 0.55 1.37 3.06

Table 10 Atherogenic indices in Spanish adolescents aged 13 to 18 years. TC: total cholesterol; HDLc: high density lipoprotein cholesterol TG: triglycerides; LDLc: low density lipoprotein cholesterol; Apo: apolipoprotein. aP < 0.05 for differences between sexes. *P < 0.05 (in comparison to girls 15 years of age).

Age groups (Years) TC/HDLc TC-HDLc (TC-HDLc)/ HDLc TG/HDLc LDLc/HDLc Apo B-100 / Apo A-1 Apo B-100 / LDLc

Males

13 3.15 2.91 2.15 0.60 1.88 0.61 0.28

14 3.04 2.70 2.04 0.63a 1.75 0.61 0.29

15 2.98 2.60 1.98 0.59a 1.71 0.58 0.29

16 2.90 2.66 1.90 0.56a 1.64 0.54 0.29a

17–18.5 3.26 2.78 2.26 0.70 1.94 0.59 0.29

Total (13–18.5) 3.06 2.73 2.06 0.62 1.78 0.58 0.29

Females

13 2.94 2.98 1.94 0.51 1.71 0.57 0.27

14 2.82 2.79 1.82 0.45 1.61* 0.63 0.29

15 2.79 2.81 1.79 0.46 1.58 0.55 0.28

16 2.76 2.70 1.76 0.45 1.56 0.52 0.28

17–18.5 2.92 2.90 1.92 0.55 1.67 0.55 0.29

Total (13–18.5) 2.85 2.83 1.85 0.48 1.63 0.56 0.28




American or Greek adolescents. Comparison of the present serum lipid profiles with those obtained in 26 other countries (Brotons et al. 1998) showed no apparent differences. Since levels of physical activity are rapidly decreasing among Span-ish adolescents (Moreno et al. 2002; Tercedor 2003) and the Mediterranean diet is losing its identity (Moreno et al. 2002; Serra-Majen et al. 1995; Zamora et al. 2003), increased obes-ity, and less favourable metabolic profile is expected to result (Moreno et al. 2002; Moreno et al. 2005). Nowadays, fruit and vegetable intake among Spanish children and adolescents is among the lowest in Europe (Yngve et al. 2005), and an increasing trend in fat consumption during the last decade has been observed (Moreno et al. 2000; Moreno et al. 2002). According to the well known relation between dietary fat, se-rum cholesterol and cardiovascular diseases (Ascherio et al. 1996), a significant increase in incidence and mortality from cardiovascular diseases should have been detected in Spain. However, this expected trend has not been observed in adults. This has been termed the ‘Spanish paradox’ (Serra-Majem et al. 1995). This paradox most likely stems from the interaction of multiple synergistic and antagonistic risk and protective factors for cardiovascular diseases.Relatively little has been published on the apolipoprotein profiles of adolescents. It is therefore difficult to compare the results of the present study with those observed in other cross-sectional examinations. Reference values for apolipoproteins in children and adolescents are of interest since they have been established as new atherosclerosis risk factor (Glowins-ka et al. 2003). According to some authors, the concentrations of apo A-1 and apo B-100 show an even stronger correlation with atheroma development than their equivalent lipoproteins HDLc and LDLc (Gomez et al. 1996). The levels seen in children have been associated with the incidence of coronary heart disease in their parents (Srinivasan & Berenson 1995). As in adults, the distribution of Lp(a) values was highly skewed towards low values. The geometric means obtained for serum Lp(a) were similar to those reported in Spain in the 1990s (Gomez et al. 1996). However, when median Lp(a) serum concentrations are compared according to age and gen-der, the figures recorded in the present study are much higher than those reported by Gomez et al. (1996). Assessing new risk factors for atherosclerosis in children and adolescents may provide new insights into the mechanism of formation of atheromatous plaques, especially during the early stages when the process is entirely reversible (Libby 2000). In this regard, the reference values for several atherogenic indices has been provide.The influence of age at the onset of menses on lipid and lipo-protein concentration is not clear. Associations between age at first menses and TG has been observed (Morrison et al. 1979),

whereas either low correlations (Freedman et al. 1987) or no ef-fect of age at first menses has been recently reported (Remsberg et al. 2005). Significant differences in lipid variables according to age of menarche were not observed in this study. This obser-vation was consistent with the above mentioned studies. The AVENA study included 2 859 adolescents, from which 581 had blood sample. The total number of adolescents to be included in the study was calculated taking into account the variance for BMI (Moreno et al. 1997), as mentioned above. Differences between BMI in the subgroup from which blood samples were obtained and the remaining subjects were not significant (Tab. 1). This suggests that the subgroup with blood data is representative of the whole population. In conclusion, the serum lipid profile of Spanish adolescents suggests that special attention should be paid to lipid status in this crucial period of life. The present study provides refer-ence data on the distribution of lipid and lipoprotein levels of Spanish adolescents, this information is crucial for planning interventions and education programs promoting the preven-tion of cardiovascular disease.

AVENA Study Group

Coordinator:A Marcos, Madrid.Principal researchers:MJ Castillo, Granada.A Marcos, Madrid.S Zamora, Murcia.M García Fuentes, Santander.M Bueno, Zaragoza.

Granada: MJ Castillo, MD Cano, R Sola, F Luyckx (Bio-chemistry), A Gutiérrez, JL Mesa, JR Ruiz (Physical fi tness),M Delgado, P Tercedor, P Chillón (Physical activity), FB Ortega, M Martín, F Carreño, GV Rodríguez, R Castillo, F Arellano (Collaborator), Universidad de Granada, E-18071 Granada.

Madrid: A Marcos, M González-Gross, J Wärnberg, S Me-dina, F Sánchez Muniz, E Nova, A Montero, B de la Rosa, S Gómez, S Samartín, J Romeo, R Álvarez, (Coordination, im-munology) A Álvarez (Cytometric analysis) L Barrios (Statis-tical analysis) A Leyva, B Payá (Psychological assessment),L Martínez, E Ramos, R Ortiz, A Urzanqui (Collaborators),Instituto de Nutrición y Bromatología, Consejo Superior de Investigaciones Científicas (CSIC), E-28040 Madrid.

Murcia: S Zamora, M Garaulet, F Pérez-Llamas, JC Baraza, JF Marín, F Pérez de Heredia, MA Fernández, C González,



R García, C Torralba, E Donat, E Morales, MD García, JA Martínez, JJ Hernández, A Asensio, FJ Plaza, MJ López (Dietanalysis), Dpto. Fisiología, Universidad de Murcia, E-30100 Murcia.

Santander: M García Fuentes, D González-Lamuño, P de Rufino, R Pérez-Prieto, D Fernández, T Amigo (Geneticstudy), Dpto. Pediatría, Universidad de Cantabria, E-19003 Santander.

Zaragoza: M Bueno, LA Moreno, A Sarriá, J Fleta, G Ro-dríguez, CM Gil, MI Mesana, JA Casajús, V Blay, MG Blay, (Anthropometric assessment), Escuela Universitaria de Cien-cias de la Salud, Universidad de Zaragoza, E-50009 Zaragoza.

Confl ict of interestNo present or past conflict of interest exists for any of the authors or their institutions.

AcknowledgmentsThis study was supported by the Spanish Ministry of Health, FEDER-FSE funds (00/0015), CSD grants 05/UPB32/01 and 09/UPB31/03, the Spanish Ministry of Education (AP2003-2128; AP-2004-2745), and grants from Panrico S.A., Madaus S.A. and Procter and Gamble S.A. We gratefully acknowledge the help of all the adolescents that took part in this study, and thank their parents and teachers for their collaboration. We also acknowledge Ms. Laura Barrios for her help with the sta-tistics, and Ms. Ulrike Albers for her help with the German.

Zusammenfassung

Referenzwerte für Serumlipide und Lipoprotein bei spani-

schen Jugendlichen: Die AVENA Studie

Ziel/Objekt: Bereitstellung aktueller Referenzwerte für Serum-

lipide und Lipoprotein spanischer Jugendlicher nach Alter und

Geschlecht

Methode: Querschnittsanalyse durchgeführt in fünf repräsen-

tativen spanischen Städten (Granada, Madrid, Murcia, Santan-

der und Zaragoza); Studienpopulation von 581 Adoleszenten

(299 Jungen und 282 Mädchen) im Alter von 13 bis 18,5 Jah-

ren. Alters- und geschlechtsspezifische Mittelwerte, Standard-

abweichungen und Perzentile wurden bestimmt für: Gesamt

(TC), Lipoprotein mit hoher Dichte (HDLc) und Lipoprotein mit

niedriger Dichte (LDLc) Cholesterol, Triglyceride, Apolipoprote-

in A-1 und B-100 und Lipoprotein (a).

Ergebnisse: Die 90igste Perzentile für TC betrug 4,95 mmol/L

in der Gruppe der Jungen und 5,19 mmol/L in der Gruppe der

Mädchen. Die HDLc-Spiegel waren in allen Altersgruppen signi-

fikant höher bei den Mädchen. Die LDLc-Werte bewegten sich

zwischen 2,32 bis 2,54 mmol/L bei den Jungen und zwischen

2,38 bis 2,62 mmol/L bei den Mädchen und waren am höchsten

bei den 13-Jährigen beider Geschlechter. Die Werte für Trigly-

ceride wiesen eine steigende Tendenz auf und waren bei den

17-Jährigen beider Geschlechter am höchsten. Die Apolipo-

protein A-1 und B-100- Spiegel entsprachen denen von HDLc

und LDLc. Der geometrische Mittelwert für Lipoprotein(a) lag

zwischen 0,44 und 0,57 μmol/L bei den Jungen und zwischen

0,50 und 0,67 μmol/L bei den Mädchen.

Fazit: Die AVENA Studie stellt Referenzmaterial von Lipiden

und Lipoprotein-Spiegeln spanischer Adoleszenter zur Verfü-

gung.

Resumé

Valeurs de référence pour les lipides et lipoprotéines sériques

chez des adolescents espagnols. l’étude AVENA

Objectives: Apporter des valeurs de référence actualisées pour

les taux sériques de lipides el lipoprotéines par rapport à l’age

et au sex.

Méthodes: Une étude transversale fût réalise en 5 villes repré-

sentatives (Granada, Madrid, Murcie, Santander et Saragosse)

incluant un échantillon représentatif de 581 adolescents (299

garçons et 282 filles), avec un age de 13 à 18.5 ans. Des moyen-

nes spécifiques pour age et sexe, avec des écarts types et

percentiles fûrent calculées pour: cholestérol total (TC), choles-

térol des lipoprotéines de haute densité (HDLc), cholestérol des

lipoprotéines de basse densité (LDLc), triglycérides, apolipo-

protéines A-I et B, et lipoprotéine (a).

Résultats: Le percentile 90 pour TC était 4.95 mmol/L pour les

garçons et 5.19 pour les filles. Les taux de HDLc étaient signi-

ficativement plus élevés chez les filles des différentes groups

d’age. Les niveaux de LDLc étaient compris entre 2.32 et

2.54 mmol/L chez les garçons, et entre 2.38 et 2.62 mmol/L chez

les filles, avec des valeurs plus élevées à 13 ans dans les deux

sexes. Les niveaux de triglycérides montraient une tendance à

augmenter progressivement jusqu’à 17 ans dans les deux sexes.

Les taux d’apolipoprotéines A-1 et B-100 étaient parallèles à

ceux de HDLc et LDLc, respectivement. La moyenne géométri-

que pour les taux de lipoprotéine (a) était comprise entre 0.44

et 0.57 μmol/L chez les garçons et entre 0.50 et 0.67 μmol/L chez

les filles.

Conclusions: Le présente étude apporte des valeurs de réfé-

rence de la distribution des taux de lipides et lipoprotéines

chez des adolescents espagnols.




References

Ascherio A, Rimm EB, Giovannucci EL, Spiegel-man D, Stampfer M, Willet WC (1996). Dietary fat and risk of coronary heart disease in men: cohort follow up study in the United States. Br Med J 313: 84–90.

Azizi F, Rahmani M, Madjid M, et al. (2001). Serum lipid levels in an Iranian population of children and adolescents: Tehran lipid and glu-cose study. Eur J Epidemiol 17: 281–8.

Berenson GS, Srinivasan SR, Bao W, Newman WP, Tracy RE, Wattigney WA (1998). Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults: the Bogalusa Heart Study. N Engl J Med 338:1650–6.

Brotons C, Ribera A, Perich RM, et al. (1998). Worldwide distribution of blood lipids and lipo-proteins in childhood and adolescence: a review study. Atherosclerosis 139: 1–9.

Freedman DS, Srinivasan SR, Webber LS, Burke GL, Berenson GS (1987). Black-white differences in serum lipoproteins during sexual maturation: the Bogalusa Heart Study. J Chronic Dis 40: 309–18.

Friedewald WT, Levy RI, Fredrickson DS (1972). Estimation of the concentration of low-density lipoprotein cholesterol in serum, without use of the preparative ultracentrifuge. Clin Chem 18:499–502.

Garcés C, Gil A, Benavente M, Viturro E, Cano B, de Oya MC (2004). Consistently High Plasma High-Density Lipoprotein-Cholesterol Levels in Children in Spain, a Country With Low Cardio-vascular Mortality. Metabolism 53: 1045–7.

Gaziano JM (1998). When should heart disease prevention begin? New Engl J Med 338: 1690–2.

Glowinska B, Urban M, Koput A, Galar M(2003). New atherosclerosis risk factors in obese, hypertensive and diabetic children and adoles-cents. Atherosclerosis 167: 275–86.

Gomez Gerique JA, Porres A, Lopez Martinez D, et al. (1996). Levels of lipoprotein(a) and serum lipids in Spanish children aged from 4 to 18 years. Acta Paediatr 85: 38–42.

González-Gross M, Castillo MJ, Moreno LA, et al. (2003a). Feeding and assessment of nutrition-al status of Spanish adolescents (AVENA study). Evaluation of risks and interventional proposal. I. Methodology. Nutr Hosp 18: 15–28.

González-Gross M, on behalf of the AVENA Group. (2003b). The experience of the AVENA multicenter study: planning and development. Ann Nutr Metab 47: 350.

Hickman TB, Briefel RR, Carroll MD, et al.(1998). Distributions and trends of serum lipid levels among United States children and adolescents ages 4–19 years: data from the Third National Health and Nutrition Examination Survey. Prev Med 27: 879–90.

Instituto Nacional de Estadística (2001). Defun-ciones según las causas de muerte más significa-tiva en España. Available at: www.ine.es

Kimm SYS, Glynn NW, Kriska AM, et al. (2002). Decline in Physical Activity in Black Girls and White Girls during Adolescence. N Engl J Med 347: 709–15.

Kwiterovich PO (1991). Serum lipid and lipopro-tein levels in childhood. Ann NY Acad Sci 623:90–107.

Libby P (2000). Changing concepts of atherogen-esis. J Intern Med 247: 349–58.

McGill HC, McMahan CA, Malcom GT, Oal-mann MC, Strong JP (1997). Effects of serum lipoproteins and smoking on atherosclerosis in young men and women. The PDAY Research Group. Pathobiological Determinants of Athero-sclerosis in Youth. Arterioscler Thromb Vasc Biol 17: 95–106.

Moreno LA, Fleta J, Mur L, Feja C, Sarría A, Bueno M (1997). Indices of body fat distribu-tion in Spanish children aged 4.0 to 14.9 years. J Pediatr Gastroenterol Nutr 25: 175–81.

Moreno LA, Mesana MI, Fleta J (2005). Overweight, obesity and body fat composition in Spanish adolescents. The AVENA Study. Ann Nut Metab 49: 71–6.

Moreno LA, Sarriá A, Lázaro A, Bueno M (2000). Dietary fat intake and body mass index in Span-ish children. Am J Clin Nutr 72 (5 supplement):1399S–403S.

Moreno LA, Sarría A, Popkin BM (2002). The nutrition transition in Spain: A European Medi-terranean country. Eur J Clin Nutr 56: 992–1003.

Morley R, Baker BA, Greene LC, Livingstone MB, Harland PS, Lucas A (1998). Dietary fibre, exercise and serum lipids and lipoprotein cholesterols in 12 to 15 year olds. Acta Paediatr 87: 1230–4.

Morrison JA, Laskarzewski PM, Rauh JL et al.(1979). Lipids, lipoproteins, and sexual matura-tion during adolescence: the Princeton maturation study. Metabolism 28: 641–9.

Nicklas B, Kaztel LI, Busby-Whitehead J, Gold-berg A (1997). Increases in high-density choles-terol with endurance exercise training are blunted in obese compared with lean men. Metabolism 46: 556–61.

Plaza Pérez y grupo de expertos de las sociedades españolas de arteriosclerosis, cardiología, pedi-atría, nutrición y medicina preventiva (1991). Report on the cholesterol levels of Spanish children and adolescents. Rev Esp Cardiol 44:567–85.

Remsberg KE, Demerath EW, Schubert CM, Chumlea WC, Sun SS, Siervogel RM (2005). Early menarche and the development of cardio-vascular disease risk factors in adolescent girls: the Fels Longitudinal Study. J Clin Endocrinol Metab 90: 2718–24.

Schulpis K, Karikas GA (1998). Serum choles-terol and triglyceride distribution in 7767 school-aged Greek children. Pediatrics 101: 861–4.

Serra-Majem L, La Vecchia C, Ribas L et al.(1993a). Changes in diet and mortality from selected cancers in southern Mediterranean countries, 1960 – 1989. Eur J Clin Nutr 47 (Suppl 1): S25–S34.

Serra-Majem L, Ribas L, Lloveras G, Salleras L(1993b). Changing patterns of fat consumption in Spain. Eur J Clin Nutr 47 (Suppl 1): S13–S20.

Serra-Majem L, Ribas L, Tresserras R, Ngo J, Salleras L (1995). How could changes in diet ex-plain changes in coronary heart disease mortality in Spain? The Spanish paradox. Am J Clin Nutr 61 (Suppl 6): 1351S–9S.

Smith SC, Jackson R, Pearson TA (2004). Principles for national and regional guidelines on cardiovascular disease prevention: a scientificstatement from the world heart and stroke forum. Circulation 109: 3112–21.

Srinivasan SR, Berenson GS (1995). Serum apolipoproteins A-I and B as markers of coronary artery disease risk in early life: the Bogalusa Heart Study. Clin Chem 41: 159–64.

Strong JP, Malcom GT, McMahan CA, et al.(1999). Prevalence and extent of atherosclerosis in adolescents and young adults: implications for prevention from the Pathobiological Determi-nants of Atherosclerosis in Youth Study. JAMA 281: 727–35.



Tercedor Sánchez P, AVENA group (2003). Physi-cal activity in adolescent as a health biomarker in adulthood [Abstract]. Ann Nutr Metab 47: 351–2.

Yngve A, Wolf A, Poortvliet E, et al. (2005). Fruit and vegetable intake in a sample of 11-year-old children in 9 European countries: The Pro Chil-dren Cross-sectional Survey. Ann Nutr Metab 49:236–45.

Zamora S, Garaulet M, Pérez de Heredia F, Pé-rez-Llamas F, AVENA Group (2003). Food habits of the Spanish adolescents in the 21st Century: Is the Mediterranean diet disappearing? [Abstract]. Ann Nutr Metab 47: 351.

Address for correspondenceJonatan R RuizDepartment of Medical Physiology School of Medicine University of Granada E-18012 Granada, Spain. Tel: 34 958 243540Fax: +34 958 246179e-mail: [email protected]

SERUM LIPIDS, BODY MASS INDEX AND WAIST

CIRCUMFERENCE DURING PUBERTAL DEVELOPMENT IN

SPANISH ADOLESCENTS; THE AVENA STUDY

Jonatan R. Ruiz1,2, Francisco B. Ortega1,2, Beatriz Tresaco3, Julia

Wärnberg2,4, José L. Mesa1, Marcela González-Gross1,5, Luis A.

Moreno3, Ascensión Marcos4, Ángel Gutiérrez1, Manuel J.

Castillo 1, and the AVENA study group

Horm Metab Res 2006; 38: 832-837

1Departamento de Fisiología, Facultad de Medicina, Universidad de Granada,

Granada, Spain, 2Unit for Preventive Nutrition, Department of Biosciences and

Nutrition at NOVUM, Karolinska Institutet, Huddinge, Sweden, 3E.U. Ciencias

de la Salud, Universidad de Zaragoza, Zaragoza, Spain, 4Grupo

Inmunonutrición, Departamento de Nutrición y Metabolismo, Consejo Superior

de Investigaciones Científicas, Madrid, Spain, 5Facultad de Ciencias de la

Actividad Física y del Deporte, Universidad Politécnica de Madrid. Spain.

II

Abstract

Aim: To describe the effects of chronological age and biological

age (pubertal development) on serum lipid and lipoprotein

levels, body mass index (BMI) and waist circumference in Span-

ish adolescents. Methods: A representative Spanish sample of

526 adolescents (254 males and 272 females), were studied.

Total cholesterol (TC), high density lipoprotein cholesterol

(HDLc), triglycerides, apolipoprotein A1 and B, and lipoprotein(a)

were measured, and low density lipoprotein cholesterol (LDLc)

was calculated. Additional measurements included BMI and

waist circumference. Adolescents were classified according to

chronological age, and pubertal development (also age of me-

narche in females). Results: In males, serum TC levels were

lower at late puberty in comparison with early puberty, and

serum LDLc levels were lower at late puberty in comparisonwith

mid and early puberty. Serum HDLc levels were lower at mid

puberty in comparisonwith early and late puberty. Serum TC and

LDLc levels were not different when analyzed according to

chronological age. In females, HDLc levels were lower at late

puberty in comparison with early and mid puberty, but no

differences were found when HDLc and the other studied lipid

and lipoprotein variables were analyzed according to chronolo-

gical age, or age of menarche. All the observed differences

persisted after adjusting for BMI and waist circumference. In

female adolescents, both BMI and waist circumference were

higher at late puberty in comparisonwith early andmid puberty,

while in males, BMI and waist circumference were different

Affiliation1Departamento de Fisiologıa, Facultad de Medicina, Universidad de Granada, Granada, Spain2Unit for Preventive Nutrition, Department of Biosciences and Nutrition at NOVUM, Karolinska Institutet,Huddinge, Sweden3 E. U. Ciencias de la Salud, Universidad de Zaragoza, Zaragoza, Spain4Grupo Inmunonutricion, Departamento de Nutricion y Metabolismo, Consejo Superior de InvestigacionesCientıficas, Madrid, Spain5 Facultad de Ciencias de la Actividad Fısica y del Deporte, Universidad Politecnica de Madrid, Spain6AVENA Study Group: Coordinator: A. Marcos, Madrid;Main Investigators: M. J. Castillo, Granada; A. Marcos,Madrid; S. Zamora, Murcia; M. Garcıa Fuentes, Santander; M. Bueno, Zaragoza, SpainGranada: M. J. Castillo, M. D. Cano, R Sola (Biochemistry); A. Gutierrez, J. L. Mesa, J. R. Ruiz (Physical fitness);M. Delgado, P. Tercedor, P. Chillon (Physical activity); F. B. Ortega, M. Martın, F. Carreno, G. V. Rodrıguez, R.Castillo, F. Arellano (Collaborators); Universidad de Granada, 18071, Granada, SpainMadrid: A. Marcos, M. Gonzalez-Gross, J. Warnberg, S. Medina, F. SanchezMuniz, E. Nova, A. Montero, B. de laRosa, S. Gomez, S. Samartın, J. Romeo, R. Alvarez, (Coordination, immunology); A. Alvarez (Cytometricanalysis); L. Barrios (Statistical analysis); A. Leyva, B. Paya (Psychological assessment); L. Martınez, E. Ramos,R. Ortiz, A. Urzanqui (Collaborators); Instituto de Nutricion y Bromatologıa, Consejo Superior de Investiga-ciones Cientıficas (CSIC), 28040 Madrid, SpainMurcia: S. Zamora, M. Garaulet, F. Perez-Llamas, J. C. Baraza, J. F. Marın, F. Perez de Heredia, M. A. Fernandez,C. Gonzalez, R. Garcıa, C. Torralba, E. Donat, E. Morales, M. D. Garcıa, J. A. Martınez, J. J. Hernandez, A. Asensio,F. J. Plaza, M. J. Lopez (Diet analysis); Dpto. Fisiologıa, Universidad de Murcia, 30100 Murcia, SpainSantander: M. Garcıa Fuentes, D. Gonzalez-Lamuno, P. de Rufino, R. Perez-Prieto, D. Fernandez, T. Amigo(Genetic study); Dpto. Pediatrıa, Universidad de Cantabria, 19003 Santander, SpainZaragoza: M. Bueno, L. A. Moreno, A. Sarria, J. Fleta, G. Rodrıguez, C. M. Gil, M. I. Mesana, J. A. Casajus, V. Blay,M. G. Blay (Anthropometric assessment); Escuela Universitaria de Ciencias de la Salud, Universidad deZaragoza, 50009 Zaragoza, Spain

CorrespondenceJonatan R. Ruiz �Departamento de Fisiologıa � Facultad de Medicina �Universidad de Granada �18012Granada � Spain �Tel.: + 34/958/24 35 40 � Fax: + 34/958/24 90 15 �E-mail: [email protected]

Received 6 April 2006 �Accepted after revision 10 July 2006

BibliographyHorm Metab Res 2006; 38: 832–837r Georg Thieme Verlag KG Stuttgart � New York �DOI 10.1055/s-2006-956503 � ISSN 0018-5043

Serum Lipids, Body Mass Index and Waist

Circumference during Pubertal Development in

Spanish Adolescents: The AVENA Study

J. R. Ruiz1,2

F. B. Ortega1,2

B. Tresaco3

J. Warnberg2,4

J. L. Mesa1

M. Gonzalez-Gross1,5

L. A. Moreno3

A. Marcos4

A. Gutierrez1

M. J. Castillo1

The AVENA Study Group6

Orig

inalClin

ical

832

when analyzed according to chronological age. Conclusion: The

results suggest that the assessment of pubertal development

may provide additional valuable information when interpreting

lipid profile and body fat in adolescents.

Key words

Adolescence � chronological age �biological age � cardiovascularrisk factors

Introduction

The pattern of changes in lipids and lipoproteins during child-

hood and adolescence have encouraged the use of age- and

gender-specific cut points for detecting children with increased,

or decreased, blood lipid levels [1, 2]. However, many factors

make the screening difficult during this period. Some previous

investigations have shown changes in serum lipids throughout

adolescence [3–6]. Serum total cholesterol (TC), low density

lipoprotein cholesterol (LDLc) and high density lipoprotein cho-

lesterol (HDLc) levels seems to decrease throughout the adoles-

cence period, which may be more readily explained by sexual

maturation rather than by chronological age. Therefore, pedia-

tricians should be aware of the influence of pubertal change on

measurements of lipoproteins. In a randomized controlled trial,

the observed lowering effect of a dietary intervention on LDLc in

children with high cholesterol was confounded by the decrease

associated with pubertal development [6]. This suggests that

during puberty, chronological age may not be an adequate dis-

criminating factor since pubertal development seems to vary

between genders and individuals [7].

Other factors such as total body fat and abdominal adiposity

have been shown to influence lipid and lipoprotein levels during

the adolescence [8, 9] and later in life [10]. We have previously

shown that both body mass index (BMI) and abdominal adipos-

ity (measured by waist circumference) are negatively associated

with lipid and lipoprotein profile in Spanish adolescents [8, 9].

The aim of this report was to describe the effects of chronological

age and biological age (pubertal development) on serum lipid

and lipoprotein levels, BMI and waist circumference in Spanish

adolescents.

Material and Methods

Study population

The subjects were participants in the AVENA (Alimentacion y

Valoracion del Estado Nutricional en Adolescentes, Food and

Nutritional Status in Adolescents) study, a cross-sectional study

designed to assess the nutritional status of a representative

sample of Spanish adolescents. The complete methodology of

the AVENA study has been described elsewhere [11–13]. The

number of subjects included in the AVENA study was 2859

adolescents. Blood samples were randomly obtained from 581

of the subjects. From these, 526 adolescents (254 males and 272

females) had a complete set of Tanner stages and lipids measure-

ments and were included in this study.

A verbal detailed description of the nature and purpose of the

study was given to adolescents and school teachers. This infor-

mation was also sent to parents or children supervisors by letter,

and the written consents from parents and adolescents were

requested. After receiving their written assent, the adolescents

were considered for inclusion in the study. Exclusion criteria

were: type 2 diabetes, pregnancy, alcohol or drug abuse, and

non-directly related nutritional medical conditions. The study

protocol was performed in accordance with the ethical standards

laid down in the 1975 Declaration of Helsinki (as revised in

Hong-Kong in 1989 and in Edinburgh in 2000), and approved

by the Review Committee for Research Involving Human Sub-

jects of the Hospital Universitario Marques de Valdecilla (San-

tander, Spain).

Physical examination

Height and weight were measured by standardized procedures.

BMI was calculated as weight/height squared (kg/m2). Waist

circumference was measured with an inelastic tape: the subject

was in a standing position, and the tape was applied horizontally

midway between the lowest rib margin and the iliac crest, at the

end of gentle expiration [14]. Technical error of measurement

was 0.95 cm, and reliability 98.0%. The technical error of mea-

surement was obtained by carrying out a number of repeated

measurements on the same subject, by the same observer; the

coefficient of reliability reveals what proportion of the between-

subject variance in a measured population is free frommeasure-

ment error [14].

Identification of pubertal stage (I–V) was assessed according to

Tanner and Whitehouse [15]. The standard staging of pubertal

maturity describes breast and pubic hair development in girls

and genital and pubic hair development in boys. There were not

any subject classified into Tanner stage I, and only 5.2 % (n = 13)

of boys and 1.7 % (n = 4) of girls were classified into Tanner stage

II. Therefore, the five established Tanner stages were re-grouped

into Tanner stage II + III, IV, and V, here called early puberty, mid

puberty and late puberty, respectively.

Age of menarche was determined from the self-reported age of

first menses based on administered questionnaire to 208 female

adolescents.

Blood sampling

Blood (20ml) was collected from an antecubital vein between

8:00 and 9:00 AM, after an overnight fast. Serum concentrations

of TC, HDLc, triglycerides (TG), apolipoprotein (apo) A1, apo B,

and lipoprotein(a) [Lp(a)] were measured. LDLc was calculated

with the Friedewald formula [16] adjusted for serum TG levels

[17]. A detailed description of the blood analysis has been

reported elsewhere [13].

Statistical analysis

Mean and standard deviation (SD) of all lipid and lipoprotein

levels were calculated according to chronological age and biolo-

gical age (pubertal development) for both male and female

adolescents, and age of menarche only for female adolescents.

Shapiro–Wilk test was used to check data distribution by gender

Ruiz JR et al. Serum Lipids and Boy Fat During Pubertal Development ... Horm Metab Res 2006; 38: 832–837

Orig

inalClin

ical

833

and age. The studied variables were quasi-normal distributed, and

the asymmetry and kurtosis levels were adequate for all, except

for Lp(a) that was achieved after logarithmic transformation.

Mean values were compared by one way analysis of covariance

(ANCOVA), and pos hoc analysis were performed by Games–Ho-

well test. Subsequent analyses were performed after adjusting

for BMI (as an index of overall corpulence) and waist circumfer-

ence (as an indicator of abdominal adiposity). BMI and waist

circumference were entered as covariates, both separately and

together. The analyses were performed using the Statistical

Package for Social Sciences (SPSS, v. 14.0 for WINDOWS; SPSS

Inc, Chicago), and the significance level was 5%.

Results

Distributions of pubertal development of study population by

age are shown in Table 1. The chronological age range of adoles-

cents within each stage of sexual maturation was large, for

example males falling in the third stage (early puberty) of

pubertal status could range from 13–18.5 years. The same is also

valid in females.

In male adolescents, serum TC levels were significantly lower at

late puberty in comparison with early puberty (Table 2). Serum

LDLc levels were significantly lower at late puberty in compar-

ison with early and mid puberty. Serum TC and LDLc levels were

not different when analysed according to chronological age

(Table 3). Serum HDLc levels were significantly lower at mid

puberty in comparison with early and late puberty (Table 2).

Serum HDLc levels were significantly lower at 17–18.5 years of

age in comparison with 15 years of age. Serum apo A1, apo B and

Lp(a) levels were not different across puberty stages, and apo A1

levels were significantly higher in males aged 16 years compared

with those aged 13, 14 and 15 years.

In female adolescents, serum HDLc levels were significantly

lower at late puberty in comparison with early and mid puberty

(Table 2). No differences were found in lipid and lipoprotein

levels according to chronological age, or age of menarche (data

not shown). All previous observed differences did not change

when the comparisons were controlled for BMI, or waist circum-

ference separately, or when both variables were entered together

as covariates.

In females, both BMI and waist circumference were significantly

higher at late puberty in comparisonwith early andmid puberty,

while no differences were observed when analyzed by chron-

ological age or age of menarche. In males, BMI was significantly

higher at 17–18.5 years of age in comparison with 13 an 16 years

of age in male adolescents. Waist circumference was signifi-

cantly higher at 17–18.5 years of age in comparison with 13

years of age in male adolescents.

The self-reported age of menarche in the present study ranged

from 9 to 15 years of age, with the following distribution: 9 years

(1.6%), 10 years (2.9%), 11 years (20.9%), 12 years (34.7%), 13 years

(27.4%), 14 years (11.3%), or 15 years (1%). Four (2%) girls reported

not to have menarche at the time of the study was performed.

Discussion

The present study describes the effects of chronological age and

biological age (pubertal development) on serum lipid and lipo-

protein levels, BMI and waist circumference in Spanish adoles-

cents. The results suggest that the assessment of pubertal

development may provide additional valuable informationwhen

interpreting lipid profile and body fat in adolescents.

Therefore, a measure of biological age should be included in

epidemiologic studies dealing with serum lipid and lipoprotein

and body fat measures among adolescents.

Our study supports previous results reporting significant effects

of pubertal development on lipid and lipoprotein levels during

adolescence [3–6,18–22]. Chronological age can be a simple

discriminating factor because it is evidently associated with

pubertal development; but, as the age of puberty onset and its

velocity vary between genders and between individuals of the

same gender [7], it represents an index not precise enough to

establish normal ranges in adolescents. Results from the present

study show that lipid distributions according to pubertal devel-

Table 1 Tanner stage distribution of study population by sex and age groups

Gender Tanner stage Age group

13 14 15 16 17–18.5

Males III (early puberty) N 22 14 7 2 5% 44.0 28.0 14.0 4.0 10.0

IV (mid puberty) N 25 21 12 23 23% 24.0 20.2 11.5 22.1 22.1

V (late puberty) N 7 17 38 20 18% 7.0 17.0 38.0 20.0 18.0

Females III (early puberty) N 6 14 0 2 2% 25.0 58.3 0.0 8.3 8.3

IV (mid puberty) N 32 25 28 38 25% 21.6 16.9 18.9 25.7 16.9

V (late puberty) N 8 12 23 41 16% 8.0 12.0 23.0 41.0 16.0


Orig

inalClin

ical

834

opment give valuable information especially in male adoles-

cents. In female adolescents, only HDLc levels were different

according to pubertal development. Subsequent analysis exam-

ining the potential effect of age at the onset of menses on lipid

and lipoprotein levels did not reveal any further information. The

influence of age at the onset of menses on lipid and lipoprotein

concentration remains to be clarified. No significant effect of age

of menarche on TC, TG, HDLc, or LDLc has been recently reported

[23], whereas others found associations between age at first

menses and TG levels [24].

Serum TC levels differed through pubertal stages in male ado-

lescents, being higher in early puberty than in late puberty,

similar to other reported studies [3–5,18]. The reported TC

decrease throughout the adolescence period is thought to be

more related to pubertal development than to chronological age.

In the present study, TC levels were not significantly different

when analyzed according to chronological age. In females ado-

lescents, the TC differences through pubertal stages seems to be

absent [4, 20], or lower than in males [3,18], which is in agree-

ment with our results. In our study, TC levels were borderline

significant, in female adolescents, however, no differences were

found when analyzed according to chronological age.

The previously reported gender differences across pubertal stages

may be due to changes in TC sub-fractions. The LDLc levels seem

to declinewith pubertal development in both genders [6]. Kwiter-

ovich and co-workers [6] found lower LDLc levels associated with

more advanced pubertal development in both boys and girls. In

our study, LDLc levels were significantly different across pubertal

development in male but not in female adolescents. The LDLc

levels were not significantly different when analyzed according to

chronological age in both male and female adolescents.

Serum HDLc levels seem to decrease with pubertal development

[3, 4, 21, 22]. In male adolescents from the AVENA study, serum

HDLc levels were lower at mid puberty in comparison with early

and late puberty. The reported differences in HDLc through

puberty seem attributable to an increase in testosterone levels

[18,19, 22]. Testosterone levels have been negatively associated

with HDLc in adolescents [3, 4, 21, 22]. Results from the Bogalusa

Heart Study [20] provided a negative relationship between tes-

tosterone and HDLc in young adolescents mainly distributed in

Tanner stages I–II. However, a positive association between

testosterone and HDLc levels, and between testosterone and

apo A1 levels was found in older adolescents who were in

advanced stages of pubertal development. These results suggest

that after completion of pubertal development (Tanner stage Vor

late puberty) the impact of endogenous testosterone on lipopro-

tein levels may be minimal, perhaps because the levels may have

exceeded a threshold. In female adolescents, the values of HDLc

did not differ across puberty stages or age at menarche, which is

in agreement with others [20, 21].

Previous studies have shown a relationship between pubertal

development and TG levels [4, 5] while others did not [3]. In our

Table 2 Lipid and lipoprotein values, bodymass index (BMI) and waist circumference (WC) in Spanish adolescents stratified by tanner stage

Outcome Males, Tanner stage

III IV V p

TC (mg/dl) 164.9 ± 23.5 158.0 ± 25.6 151.9 ± 28.1� 0.017

HDLc (mg/dl) 53.8 ± 9.5 48.3 ± 11.9�� 52.6 ± 11.1 < 0.001

LDLc (mg/dl) 97.1 ± 21.3 94.5 ± 22.9 86.0 ± 23.9�� 0.006

TG (mg/dl) 70.2 ± 27.6 76.0 ± 30.4 66.4 ± 39.8 0.168

Apo A1 (mg/dl) 118.1 ± 22.2 112.6 ± 23.3 113.8 ± 17.0 0.175

Apo B100 (mg/dl) 69.3 ± 13.0 68.4 ± 14.1 65.8 ± 14.6 0.289

Lp (a) (mg/dl) 13.0 ± 5.2 12.9 ± 4.4 15.0 ± 3.9 0.746

BMI (kg/m2) 21.7 ± 4.5 22.7 ± 4.3 21.8 ± 3.3 0.180WC (cm) 75.8 ± 10.9 78.1 ± 10.2 76.8 ± 8.4 0.334

Outcome Females, Tanner stage

III IV V p

TC (mg/dl) 176.8 ± 23.5 171.9 ± 25.6 168.0 ± 28.1 0.090

HDLc (mg/dl) 62.7 ± 9.5 60.8 ± 11.9 56.3 ± 11.1] 0.006

LDLc (mg/dl) 101.8 ± 21.3 97.7 ± 22.9 98.4 ± 23.9 0.687

TG (mg/dl) 61.3 ± 27.6 67.3 ± 30.4 67.0 ± 39.8 0.577

Apo A1 (mg/dl) 123.8 ± 22.2 125.6 ± 23.3 119.0 ± 17.0 0.262

Apo B100 (mg/dl) 71.6 ± 13.0 70.9 ± 14.1 71.7 ± 14.6 0.891

Lp (a) (mg/dl) 19.7 ± 2.9 14.9 ± 4.0 16.3 ± 4.4 0.650

BMI (kg/m2) 19.9 ± 3.1 21.2 ± 3.0 22.6 ± 4.04 0.001

WC (cm) 65.8 ± 7.2 69.7 ± 7.1 73.8 ± 8.344 < 0.001

Values are means ± SD. TC: total cholesterol; HDLc: high density lipoprotein cholesterol; LDLc: low density lipoprotein cholesterol; TG: triglycerides; Apo:apolipoprotein; Lp(a): lipoprotein a.Geometric mean ± SD. �p = 0.019 in comparison to Tanner stage III. ��p = 0.019 and 0.018 in comparison toTanner stage III and V, respectively. ��p = 0.013 and 0.015 in comparison to Tanner stage III and IV, respectively. ]p = 0.013 and 0.019 in comparisonto Tanner stage II and IV, respectively.4p = 0.004 and 0.0035 in comparison to Tanner stage III and IV, respectively.44p < 0.001 and 0.002 in comparison toTanner stage III and IV, respectively.


Orig

inalClin

ical

835

study, TG levels did not differ across puberty stages neither in

male nor in female adolescents (Table 2). Although TG levels have

been suggested to be better explained by chronological age than

by pubertal development in males [3, 4], we did not find differ-

ences when analyzed according to chronological age (Table 3).

Relatively little has been published about the apolipoprotein

profiles in adolescents. Serum apolipoproteins and Lp(a) levels

were not different when analyzed according to pubertal devel-

opment, neither when analyzed according to chronological age,

except for apo A1 levels, nor when analyzed according to chron-

ological age in males (Table 3). Serum levels of Lp(a) were not

different either when analyzed according to chronological age or

according to pubertal development, as it has been reported ear-

lier [4]. This supports the contention that Lp(a) is predominantly

genetically controlled.

One interesting finding is that apo B and A1 levels were not

concordant with those for LDLc and HDLc, respectively. One

possible explanation to the absence of concordance of LDLc and

apo B levels may be because LDLc particles are losing cholesterol.

When LDL particles lose cholesterol, the particle becomes smal-

ler. Moreover, if the LDL particle loses cholesterol but not apo B

which means that the LDL particle is becoming smaller and

denser. The same apply to HDLc particle. According to several

cross-sectional and prospective epidemiological studies, sub-

jects with small, dense LDLc particles have a higher risk of

coronary artery disease than subjects with large, buoyant LDLc

particles [25]. Possible mechanisms mediating this increased

atherogenicity of small LDLc particles include increased oxida-

tion, diminished binding affinity to LDLc receptors [26], in-

creased binding to arterial wall proteoglycans [27], and

impaired in vivo endothelial function independent of HDLc, LDLc

and TG concentrations. Reference values for the apo B and LDLc

ratio in adolescents have recently been published [13].

All the above observed differences did not change when the

comparisons were controlled for BMI and waist circumference,

which may suggest that the associations between sexual ma-

turation and lipid and lipoprotein profile are independent of

body composition and body fat distribution at these ages.

The findings that both BMI andwaist circumferencewere different

across pubertal stages in females, but were not when analyzed

according to chronological age or age of menarche are in concor-

dance with others [28, 29]. Similarly, no association between BMI

and age of menarche has been recently reported in a prospective

study involving 124 healthy girls aged 8 to 18 years [28].

Taking together, these results suggest that pubertal development

seems to have an influence on lipid and lipoprotein profile and

body composition in adolescents. These findings may be related

to disparate hormonal patterns that emerge during adolescence.

Adolescence is highly sensitive to environmental factors which

may influence the endogenous hormonal milieu. However, we

did not measure sex hormones, which hamper a further study of

hormone-lipoprotein relationships in the studied population.

Table 3 Lipid and lipoprotein values, body mass index (BMI) and waist circumference (WC) in Spanish adolescents stratified by age group

Outcome Male, age group

13 14 15 16 17–18.5 p

TC (mg/dl) 164.5 ± 31.0 155.3 ± 22.9 151.0 ± 23.0 157.1 ± 24.8 155.0 ± 28.0 0.102

HDLc (mg/dl) 52.2 ± 11.0 51.0 ± 10.4 50.7 ± 8.9 54.3 ± 10.3 47.5 ± 6.9� 0.034

LDLc (mg/dl) 97.9 ± 25.3 89.5 ± 20.7 86.5 ± 21.0 88.8 ± 22.7 92.2 ± 27.5 0.133

TG (mg/dl) 72.2 ± 36.6 74.1 ± 36.0 68.9 ± 25.1 70.2 ± 29.2 76.5 ± 32.5 0.794

Apo A1 (mg/dl) 115.7 ± 17.0 110.3 ± 19.4 111.6 ± 19.5 126.1 ± 20.2�� 119.6 ± 17.2 0.011

Apo B100 (mg/dl) 70.9 ± 18.4 66.8 ± 15.3 64.9 ± 14.1 67.5 ± 13.3 70.3 ± 15.8 0.335

Lp(a) (mg/dl) 12.3 ± 1.6 13.8 ± 1.4 13.7 ± 1.2 13.5 ± 1.6 15.9 ± 1.7 0.653

BMI (kg/m2) 20.6 ± 3.2 22.0 ± 3.9 22.2 ± 4.0 21.8 ± 3.5 24.2 ± 4.5�� < 0.001

WC (cm) 74.3 ± 9.4 77.4 ± 10.7 77.1 ± 9.2 76.2 ± 7.6 81.0 ± 10.3] 0.001

Outcome Females, age group

13 14 15 16 17–18.5 p

TC (mg/dl) 174.3 ± 22.9 166.9 ± 27.2 169.2 ± 24.5 163.2 ± 26.8 170.0 ± 29.2 0.229

HDLc (mg/dl) 59.2 ± 10.5 59.2 ± 10.8 66.6 ± 12.7 59.0 ± 12.5 58.1 ± 10.8 0.258

LDLc (mg/dl) 101.3 ± 20.2 95.6 ± 25.9 95.8 ± 21.6 91.9 ± 22.5 97.1 ± 25.1 0.283

TG (mg/dl) 68.6 ± 21.1 60.6 ± 24.2 63.9 ± 20.2 61.4 ± 21.1 73.9 ± 56.4 0.108

Apo A1 (mg/dl) 124.4 ± 15.3 114.3 ± 25.8 127.6 ± 24.0 129.5 ± 22.8 133.8 ± ± 21.7 0.110

Apo B100 (mg/dl) 71.4 ± 11.2 71.6 ± 16.4 70.1 ± 13.4 67.6 ± 13.3 73.1 ± 15.1 0.316

Lp(a) (mg/dl) 16.6 ± 1.4 14.7 ± 1.5 15.4 ± 1.3 13.9 ± 1.5 18.77 ± 1.4 0.990

BMI (kg/m2) 21.0 ± 3.9 21.7 ± 4.2 21.3 ± 3.2 22.3 ± 3.0 21.7 ± 2.7 0.581

WC (cm) 70.3 ± 9.0 70.7 ± 10.7 70.8 ± 6.7 72.2 ± 7.7 70.1 ± 6.0 0.834

Values are means ± SD. TC: total cholesterol; HDLc: high density lipoprotein cholesterol; LDLc: low density lipoprotein cholesterol; TG: triglycerides; Apo:apolipoprotein; :p(a): lipoprotein a.Geometric mean ± SD. �p = 0.016 in comparison to 15 years of age. ��p = 0.025, 0.013, 0.015 in comparison to 13, 14and 15 years of age, respectively. ��p < 0.001 and 0.039 in comparison to 13 and 16 years of age, respectively. ]p < 0.08 in comparison to 13 years of age.


Orig

inalClin

ical

836

The results should be interpreted with caution due to the limita-

tions of the cross-sectional nature of the study. The absence of

adolescents in Tanner stage I and II limits the possibility to make

comparisons over the full spectrumof pubertal development. Long-

itudinal studies are needed in order to accurately examine the

tracking of lipid and lipoprotein levels over adolescence, to accu-

rately study the age- and pubertal development-related changes

during this important period of life. The measurement of apolipo-

proteins and the fact that the present study sample is representa-

tive of the whole population [13] are strengths of the study.

In conclusion, results from this study suggest that the assess-

ment of pubertal development may provide additional valuable

information when interpreting lipid profile and body fat in

adolescents.

Acknowledgements

This study was supported by the Spanish Ministry of Health (00/

0015) and FEDER-FSE funds, CSD (05/UPB320, and 109/UPB31/

0313/UPB20/04), Spanish Ministry of Education (AP2002-2920,

AP2003-2128; and AP-2004-2745), and grants from Panrico S.A.,

Madaus S.A. and Procter and Gamble S.A. We gratefully acknowl-

edge all participating adolescents, and their parents and teachers

for their collaboration. We also acknowledge Ms Laura Barrios

for her statistical support.

References

1 Dennison BA, Kikuchi DA, Srinivasan SR, Webber LS, Berenson GS.Serum total cholesterol screening for the detection of elevated low-density lipoprotein in children and adolescents: The Bogalusa HeartStudy. Pediatrics 1990; 85: 472–479

2 Porkka KVK, Viikari JSA, Ronnemaa T, Marniemi J, Akerblom HK. Ageand gender specific serum lipid and apolipoprotein fractiles of Fin-nish children and young adults: The Cardiovascular Risk in YoungFinns Study. Acta Paediatr 1994; 83: 838–448

3 Tell GS, Mittelmark MB, Vellar OD. Cholesterol, high density lipopro-tein cholesterol and triglycerides during puberty: The Oslo YouthStudy. Am J Epidemiol 1985; 122: 750–761

4 Cobbaert C, Deprost L, Mulder P, Rombaut K, Gijsels G, Kesteloot H.Pubertal serum lipoprotein (a) and its correlates in Belgian school-children. Int J Epidemiol 1995; 24: 78–87

5 Bertrais S, Balkau B, Charles MA, Vol S, Calvet C, Tichet J et al. Puberty-associated differences in total cholesterol and triglyceride levelsaccording to sex in French children aged 10–13 years. Ann Epidemiol2000; 10: 316–323

6 Kwiterovich PO, Barton BA, McMahon RP, Obarzanek E, Hunsberger S,Simons-Morton D et al. Effects of diet and sexual maturation on low-density lipoprotein cholesterol during puberty: The Dietary Interven-tion Study in Children (DISC). Circulation 1997; 96: 2526–2533

7 Preece MA. Evaluation of growth and development. In: Avner, ED,Barratt TM, Holliday MA, eds Pediatric Nephrology 3rd ed. Philadel-phia, PA, Williams & Wilkins, 1994; 37–396

8 Mesa JL, Ruiz JR, Ortega FB, Warnberg J, Gonzalez-Lamuno D, MorenoLA et al. Aerobic physical fitness in relation to blood lipids and fastingglycaemia in adolescents Influence of weight Status. Nutr MetabCardiovasc Dis 2006; 16: 285–293

9 Mesa JL, Ortega FB, Ruiz JR, Castillo MJ, Tresaco B, Carreno F et al.Anthropometric determinants of a clustering of lipid-related ather-ogenesis risk factors in overweight and non-overweight adolescentsInfluence of cardiorespiratory fitness. Ann Nutr Metab in press

10 Berg G, Mesch V, Boero L, Sayegh F, Prada M, Royer M et al. Lipid andlipoprotein profile in menopausal transition Effects of hormones, ageand fat distribution. Horm Metab Res 2004; 36: 215–220

11 Gonzalez-Gross M, Castillo MJ, Moreno LA, Nova E, Gonzalez-LamunoD, Perez-Llamas F et al. Alimentacion y valoracion del estado nutri-cional de los adolescentes espanoles (Estudio AVENA) Evaluacion deriesgos y propuesta de intervencion. I. Descripcion metodologica delproyecto. Nutr Hosp 2003; 18: 15–28

12 Moreno LA, Mesana MI, Fleta J, Ruiz JR, Gonzalez-Gross M, Sarrıa A,Marcos A, Bueno M, AVENA Study Group. Overweight, obesity andbody fat composition in Spanish adolescents The AVENA Study. AnnNut Metab 2005; 49: 71–76

13 Ruiz JR, Ortega FB, Moreno LA, Warnberg J, Gutierrez A, Gonzalez-Gross, Cano MD, Gutierrez A, Castillo MJ. Reference values for serumlipids and lipoproteins in Spanish adolescents. The AVENA study. SozPraventiv Med 2006; 51: 99–109

14 Moreno LA, Joyanes M, Mesana MI, Gonzalez-Gross M, Gil CM, SarriaA, Gutierrez A, Garaulet M, Perez-Prieto R, Bueno M, Marcos A, AVENAStudy Group. Harmonization of anthropometric measurements for amulticenter nutrition survey in Spanish adolescents. Nutrition 2003;19: 481–486

15 Tanner JM, Whitehouse RH. Clinical longitudinal standards for height,weight, height velocity and stages of puberty. Arch Dis Child 1976; 51:170–179

16 Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentra-tion of low-density lipoprotein cholesterol in serum, without use ofthe preparative ultracentrifuge. Clin Chem 1972; 18: 499–502

17 Morley R, Baker BA, Greene LC, Livingstone MB, Harland PS, Lucas A.Dietary fibre, exercise and serum lipids and lipoprotein cholesterols in12 to 15 year olds. Acta Paediatr 1998; 87: 1230–1234

18 Berenson GS, Srinivasan SR, Cresanta JL, Foster TA, Webber LS. Dy-namic changes of serum lipoproteins in children during adolescenceand sexual maturation. Am J Epidemiol 1981; 113: 157–170

19 Dacou-Voutetakis C, Trichopoulou A, Papazissis G, Lagos P, Clontza D,Agathopoulos A et al. Serum cholesterol concentration in Greek pupilsaged 9–18 years The effect of puberty. Acta Paediatr Scand 1981; 70:257–258

20 Srinivasan SR, Freedman DS, Sundaram GS, Webber LS, Berenson GS.Racial (black-white) comparisons of the relationship of levels ofendogenous sex hormones to serum lipoproteins during male adoles-cence: the Bogalusa Heart Study. Circulation 1986; 74: 1226–1234

21 Srinivasan SR, Wattigney W, Webber LS, Berenson GS. Race andgender differences in serum lipoproteins of children, adolescents,and young adults-Emergence of an adverse lipoprotein pattern inwhite males: The Bogalusa Heart Study. Prev Med 1991; 20: 671–684

22 Morrison JA, Barton BA, Biro FM, Sprecher DL. Sex hormones and thechanges in adolescent male lipids: longitudinal studies in a biracialcohort. J Pediatr 2003; 142: 637–642

23 Remsberg KE, Demerath EW, Schubert CM, Chumlea WC, Sun SS,Siervogel RM. Early menarche and the development of cardiovasculardisease risk factors in adolescent girls: the Fels Longitudinal Study. JClin Endocrinol Metab 2005; 90: 2718–2724

24 Morrison JA, Laskarzewski PM, Rauh JL, Brookman R, Mellies M, FrazerM, Khoury P, deGroot I, Kelly K, Glueck CJ. Lipids, lipoproteins, andsexual maturation during adolescence: the Princeton maturationstudy. Metabolism 1979; 28: 641–649

25 Lamarche B, Tchernof S, Moorjani B, Cantin GR, Dagenais PJ, Lupien PJet al. Small, dense low-density lipoprotein particles as a predictor ofthe risk of ischemic heart disease inmen: prospective results from theQuebec Cardiovascular Study. Circulation 1997; 95: 69–75

26 Glomset JA, Assmann G, Gjone E, Norum KR. Lecithin:cholesterolacyltransferase deficiency and fish-eye disease. In: Scriber CR, Beau-det AL, Sly WS, Valle D, eds. The Metabolic and Molecular Basis ofInherited Disease 7th ed, Vol. II. New York: McGraw-Hill, Inc., 1995,1993–1951

27 Anber V, Griffin BA, McConnell M, Packard CJ, Shepherd J. Influence ofplasma lipid and LDL-subfraction profile on the interaction betweenlow density lipoprotein with human arterial wall proteoglycans.Atherosclerosis 1996; 124: 261–271

28 Rapkin AJ, Tsao JC, Turk N, Anderson M, Zeltzer LK. Relationshipsamong Self-Rated Tanner Staging, Hormones, and Psychosocial Fac-tors in Healthy Female Adolescents. J Pediatr Adolesc Gynecol 2006;19: 181–187

29 Hoffman WH, Barbeau P, Litaker MS, Johnson MH, Howe CA, Gutin B.Tanner staging of secondary sexual characteristics and body compo-sition, blood pressure, and insulin in black girls. Obes Res 2005; 13:2195–2201


Orig

inalClin

ical

837

HEALTH-RELATED FITNESS ASSESSMENT IN CHILDHOOD

AND ADOLESCENCE: A EUROPEAN APPROACH BASED ON

THE AVENA, EYHS AND HELENA STUDIES

Jonatan R. Ruiz1,2, Francisco B. Ortega1,2, Ángel Gutiérrez1, Dirk

Meusel3, Michael Sjöström2, Manuel J. Castillo 1

J Public Health 2006; 14: 269-277

1 Department of Physiology, School of Medicine, University of Granada,


Nutrition at NOVUM, Karolinska Institutet, Huddinge, Sweden, 3Research

Association Public Health, Institute of Clinical Pharmacology,

Medical Faculty, Dresden University of Technology, Germany.

III

REVIEW ARTICLE

Health-related fitness assessment in childhoodand adolescence: a European approach basedon the AVENA, EYHS and HELENA studies

Jonatan R. Ruiz & Francisco B. Ortega &

Angel Gutierrez & Dirk Meusel & Michael Sjöström &

Manuel J. Castillo

Received: 15 June 2006 /Accepted: 22 June 2006 / Published online: 21 September 2006# Springer-Verlag 2006

Abstract Results from cross-sectional and longitudinalstudies such as Alimentación y Valoración del EstadoNutricional en Adolescentes: Food and Assessment of theNutritional Status of Spanish Adolescents (AVENA) andthe European Youth Heart Study (EYHS) respectively,highlight physical fitness as a key health marker inchildhood and adolescence. Moderate and vigourous levelsof physical activity stimulate functional adaptation of alltissues and organs in the body (i.e. improve fitness),thereby also making them less vulnerable to lifestyle-related degenerative and chronic diseases. To identifychildren and adolescents at risk for these major publichealth diseases and to be able to evaluate the effects ofalternative intervention strategies in European countries andinternationally, comparable testing methodology across

Europe has to be developed, tested, agreed upon andincluded in the health monitoring systems currently underdevelopment by the European Commission (EC): theDirectorate General for Health and Consumer Affairs (DGSANCO); the Statistical Office of the European Communi-ties (EUROSTAT), etc. The Healthy Lifestyle in Europe byNutrition in Adolescence (HELENA) study group plans,among other things, to describe the health-related fitness ofadolescents in a number of European countries. Experi-ences from AVENA and EYHS will be taken advantage of.This review summarises results and experiences from thedevelopmental work so far and suggests a set of health-related fitness tests for possible use in future healthinformation systems.

Keywords Cardiorespiratory fitness . Muscular fitness .

Physical activity . Non-communicable diseases .

Young adults . Health-related fitness

Introduction

The public health burden of lifestyle-related diseases in theEuropean countries is high. The most common causes ofmorbidity and mortality are coronary heart disease, stroke,obesity, hypertension, type-2 diabetes, allergies and severalcancers. A sedentary lifestyle is a major risk factor for thesediseases and is close to overtaking tobacco as the leadingcause of preventable death (Mokdad et al. 2004). Theprotective effect of intentional physical activity on theabove mentioned non-communicable diseases has beenwidely reported in people of all ages (Strong et al. 2005;Jonker et al. 2006). Regular participation in moderate andvigorous levels of exercise increases physical fitness, whichcan lead to many health benefits (Ruiz et al. 2006a).

J Public Health (2006) 14:269–277DOI 10.1007/s10389-006-0059-z

On behalf of the HELENA Study Group

J. R. Ruiz : F. B. Ortega :A. Gutierrez :M. J. CastilloDepartment of Physiology, School of Medicine,University of Granada,Granada, Spain

D. MeuselResearch Association Public Health Institute of ClinicalPharmacology, Medical Faculty,Technische Universität,Dresden, Germany

J. R. Ruiz : F. B. Ortega :M. SjöströmUnit for Preventive Nutrition,Department of Biosciences and Nutrition at NOVUM,Karolinska Institutet,Huddinge, Sweden

M. J. Castillo (*)Department of Physiology, School of Medicine,University of Granada,Granada, Spaine-mail: [email protected]

Physical fitness is also determined by constitutional factors,and it has been suggested that up to ∼40% of variation infitness may be attributable to genetic factors (Bouchard1986). In adults, low physical fitness (mainly low cardio-respiratory fitness and low muscular strength) seems to be astronger predictor of both cardiovascular and all-causemortality than any other well established risk factors(Myers et al. 2002). In Spanish adolescents, results fromthe Alimentación y Valoración del Estado Nutritional enAdolescentes: Food and Assessment of the NutritionalStatus of Spanish Adolescents (AVENA) study; (http://www.estudioavena.com), suggest significant associationsbetween cardiorespiratory fitness and plasma lipid profile(Mesa et al. 2006a) inflammatory status (Wärnberg 2006)and abdominal adiposity (Ortega et al. in press). Similarresults have been achieved in Swedish and Estonianchildren aged 9–10 years from the European Youth HearthStudy (EYHS), as well as in other cross-sectional andlongitudinal studies across Europe (Ruiz et al. 2006a,b).Taken together, these results may have important impli-cations for public-health-oriented lifestyle interventionprograms.

Physical fitness refers to the full range of physicalqualities, i.e. cardiorespiratory fitness, muscular strength,speed of movement, agility, coordination, and flexibility. Itcan be understood as an integrated measurement of allfunctions (skeletomuscular, cardiorespiratory, haematocir-culatory, psychoneurological and endocrine–metabolic) andstructures involved in the performance of physical activityand/or physical exercise (Castillo Garzon et al. 2005).There are several well-known, health-related fitness bat-teries to assess fitness in all its dimensions in young people.A good example in Europe is the EUROFIT battery(Committee of Experts on Sports Research EUROFIT,1993) and in the USA is the FITNESSGRAM battery

(Cooper Institute for Aerobics Research 1999). A numberof studies have followed most of the indications given inthese and other fitness batteries. Some of the suggestedhealth-related fitness tests have been performed in Amer-ican (Baquet et al. 2006), Finnish (Mikkelsson et al. 2006),Russian (Izaak and Panasiuk 2005), Greek (Koutedakis andBouziotas 2003), Flemish (Deforche et al. 2003), African(Monyeki et al. 2005), Spanish (Ortega et al. 2005), Dutch(Kemper et al. 2000) and Swedish and Estonian (Ruiz et al.2006a,b) adolescents. However, in most studies, an adap-tation of the tests has been made according to local/nationalsocial, cultural or environmental considerations and instru-ment or budget issues at the time the study was done.

To identify children and adolescents at risk for the majorpublic health diseases and to be able to evaluate effects ofalternative intervention strategies in European countries andinternationally, comparable testing methodology acrossEurope has to be developed, tested, agreed upon andincluded in the health monitoring systems currently underdevelopment by the European Commission (EC) (DGSANCO; EUROSTAT, etc.). In this work, experiences fromprevious projects across Europe (AVENA and EYHS) willbe taken advantage of. The Healthy Lifestyle by Nutritionin Adolescence (HELENA) study; (http://www.helenastudy.com) is a European-Union (EU)-funded project on lifestyleand obesity among European adolescents. The HELENAstudy will provide, for the first time in Europe, harmonisedand comparable data about health-related fitness and otherhealth-related outcomes among male and female adoles-cents from ten European countries (Athens in Greece,Dortmund in Germany, Gent in Belgium, Heraklion inCrete, Lille in France, Pecs in Hungary, Rome in Italy,Stockholm in Sweden, Vienna in Austria and Zaragoza inSpain). The health-related fitness test battery suggested forthe HELENA study is summarised in Table 1. Methods for

Table 1 Summary of health-related fitness tests included in the Healthy Lifestyle in Europe by Nutrition in Adolescence (HELENA) study

Fitnessdimensions

Fitness quality Test Included in theEUROFIT battery

Included in theFITNESSGRAM battery

Cardiorespiratoryfitness

Aerobic capacity 20-m shuttle run ✓ ✓

Flexibility Flexibility Back-saver sit and reach ✓

Muscular fitness Maximal isometric musclestrength

Handgrip strength ✓

Muscular endurance Curl up ✓

Explosive strength Standing broad jump ✓ ✓

Explosive strength, elasticenergy, coordination

Squat jump, counter movementjump, Abalakov

Muscular endurance Bent-arm hang ✓ ✓

Speed ofmovement–agility

Speed, agility andcoordinationa

Shuttle run 4×10-m ✓

aModified from the EUROFIT battery

270 J Public Health (2006) 14:269–277

health-related fitness assessment have already been testedfor feasibility and reliability.

This review summarises results and experiences from thedevelopmental work so far in AVENA, EYHS andHELENA studies and suggests a set of health-relatedfitness tests for possible use in future health informationsystems.

Assessment of cardiorespiratory fitness

What is cardiorespiratory fitness?

Cardiorespiratory fitness is one of the most importantcomponents of health-related fitness. Cardiorespiratoryfitness reflects the overall capacity of the cardiovascularand respiratory systems and the ability to carry outprolonged strenuous exercise. Hence, cardiorespiratoryfitness has been considered a direct measure of thephysiological status of the person. Cardiorespiratory fitness,cardiovascular fitness, cardiorespiratory endurance, aerobicfitness, aerobic capacity, aerobic power, maximal aerobicpower, aerobic work capacity, physical work capacity andmaximal oxygen consumption (VO2max) all refer to thesame concept and are used interchangeably in the literature.

In this manuscript, only the term cardiorespiratory fitness isused.

Why is cardiorespiratory fitness important in the youngpopulation?

High cardiorespiratory fitness during childhood and ado-lescence has been associated with a healthier cardiovascularprofile during these years (Mesa et al. 2006a,b) and later inlife (for review see Ruiz et al. 2006a,b). Results from theSwedish and Estonian part of the EYHS revealed negativeassociations between cardiorespiratory fitness and body fat(expressed as the sum of five skin folds) (Ruiz et al. 2006a).The same relationship was noted between cardiorespiratoryfitness and other features of the metabolic syndrome[insulin resistance, raised triglycerides and total cholesterolto high-density lipoprotein (HDL) cholesterol ratio] inchildren (Ruiz et al. 2006b). Similar results have beenfound in Spanish counterparts from the AVENA study(Gonzalez-Gross et al. 2003; Mesa et al. 2006a) (Fig. 1). Inthe same study, we have shown associations betweenincreased cardiorespiratory fitness and a favourable meta-bolic profile in both overweight and non-overweightadolescents [normal-weight category was categorised fol-

Fig. 1 Physical fitness variablesassociated with cardiovascularrisk factors among normal-weight Spanish adolescents.Normal-weight category wascategorised following the Inter-national Obesity Task Force(IOTF)-proposed gender- andage-adjusted body mass index(BMI) cutoff points

J Public Health (2006) 14:269–277 271

lowing the International Obesity Task Force (IOTF)-proposed gender- and age-adjusted body mass index(BMI) cutoff points (Cole et al. 2000)], and the mainoutcome was that cardiorespiratory fitness was an indicatorof a favourable metabolic profile in male adolescents (Mesaet al. 2006a). Results are similar in other European childrenand adolescents (Klasson-Heggebo et al. 2006).

A number of longitudinal studies have suggested thatlow cardiorespiratory fitness during childhood and adoles-cence is associated with later cardiovascular risk factors,such as hyperlipidemia, hypertension and obesity (forreview, see Ruiz et al. 2006b).

Cardiorespiratory fitness test methodology in young people

One of the most widely used tests to assess cardiorespira-tory fitness among children and adolescents is the 20-mshuttle run test, also called “Course Navette” test (Léger etal. 1984). The initial speed is 8.5 km/h, which is increasedby 0.5 km/h per min (1 min equal to one stage). Subjectsare instructed to run in a straight line, to pivot uponcompleting a shuttle, and to pace themselves in accordancewith audio signals given. The test is finished when thesubject failed to reach the end lines concurrent with theaudio signals on two consecutive occasions. A moredetailed methodology and reference values of ∼3,000Spanish adolescents participating in the AVENA studycan be found elsewhere (Ortega et al. 2005). The equationsof Leger et al. (1984) are used to estimate the VO2max fromthe result of the 20-m shuttle run test: VO2max=31.025+3.238S−3.248A+0.1536SA, where A is the age and S thefinal speed (S=8+0.5 x last stage completed). Reliabilityand validity of this test for determining the VO2max inchildren and adolescents has been widely documented. Thetest has many advantages as a fitness test because a largenumber of subjects can be tested at the same time, whichenhances participant motivation and, because of its objec-tivity, standardisation, reliability, validity and availability ofreference data. The 20-m shuttle run test has been includedin several fitness batteries, such as the EUROFIT (Com-mittee of Experts on Sports Research EUROFIT 1993), theAustralian Coaching Council (Australian Sports Commis-sion 1999), the British National Coaching Foundation(Brewer et al. 1988), the American Progressive AerobicCardiovascular Endurance Run (Cooper Institute for Aero-bics Research 1999), and the Queen’s University (Riddoch1990), among others.

Previous cross-sectional and longitudinal Europeanstudies (e.g. EYHS) have used a maximum cycle ergometertest (Hansen et al. 1989). This test is probably one of themost objective, reliable and valid indicator of cardiorespi-ratory fitness, but it is demanding on resources, especiallywhen large groups of subjects are tested. Moreover, a major

limitation to cycle ergometer testing is the discomfort andfatigue of the muscle quadriceps. In inexperienced subjects,leg fatigue may cause him/her to stop before reaching a trueVO2max. There are some studies showing that VO2max, theventilatory threshold, and minute ventilation are generally10–20% higher with treadmill testing (Working Group onCardiac Rehabilitation and Exercise Physiology 2001).

Assessment of flexibility

What is flexibility?

Flexibility is the ability of a specific muscle or musclegroup to move freely through a full range of motion. It is ofimportance in a variety of athletic performances but also inthe capacity to carry out the activities of daily living, whichis very important from a public health perspective.

“Back-saver sit-and-reach”

What is “back-saver sit-and-reach?”

Back-saver sit and reach assesses flexibility by means ofreaching forward as far as possible from a seated positionwith one leg bent at knee. The test requires a standardisedbox with a ruler, which has to be pushed by the subject.

Why is performing “back-saver sit-and-reach” importantin the young population?

There is growing evidence about the associated benefits offlexibility, including range of motion and function, im-proved athletic performance, reduced injury risk, preven-tion or reduction of postexercise soreness and improvedcoordination (Pope et al. 2000). Some studies have shownthat decreased hamstring flexibility is a risk factor for thedevelopment of patella tendinopathy and patellofemoralpain (Witvrouw et al. 2000, 2001), hamstring strain injury(Witvrouw et al. 2001) and symptoms of muscle damagefollowing eccentric exercise (McHugh et al. 1999). Simi-larly, poor flexibility and subsequent injury has beenestablished in several musculotendinous units, includingthe Achilles tendon (Leach et al. 1981) and plantar fascia(Kibler et al. 1991). Results from a recent longitudinalFinnish study suggest that hamstring flexibility (measuredby the sit-and-reach test) was one of the best explanatoryfactors for adult health-related fitness for men (Mikkelssonet al. 2006).

Back-saver sit-and-reach test methodology in the young

One of the tests to assess lower body flexibility is the back-saver sit-and-reach test. The back-saver sit-and-reach test is

272 J Public Health (2006) 14:269–277

part of the FITNESSGRAM battery (Cooper Institute forAerobics Research 1999), and is a modification of the moretraditional sit-and-reach test included in the EUROFITbattery (Committee of Experts on Sports Research EURO-FIT 1993). The back-saver sit-and-reach test differs fromthe sit-and-reach test in that the subject performs the testwith one leg bent at the knee; therefore, it may be safer onthe back by restricting flexion. The traditional sit-and-reachtest (both legs are stretched simultaneously) may result inoverstretching of the lower back, especially in terms ofexcessive disc compression and posterior ligament anderector spinae muscle strain. It also involves a forwardrotation of the pelvis and sacrum which elongates thehamstrings. The back-saver sit-and-reach allows the legs tobe evaluated separately and therefore also the determinationof symmetry (or asymmetry) in hamstring flexibility. Inaddition, testing one leg at a time eliminates the possibilityof hyperextension of both knees. The reliability and validityof the back-saver sit-and-reach tests has been widelyreported (Cooper Institute for Aerobics Research 1999).The sit-and-reach test has been usually performed in thebackground of school physical education classes, suggestingits feasibility and applicability in this context. Therefore, thepossibility of preforming the back-saver sit-and-reach testinstead of sit-and-reach test would not be a problem.

Assessment of muscular fitness

Balanced, healthy functioning of the musculoskeletalsystem requires that a specific muscle or muscle group beable to generate force or torque (measured as strength),resist repeated contractions over time or maintain amaximal voluntary contraction for a prolonged period oftime (measured as muscular endurance) and to carry out amaximal, dynamic contraction of a muscle or muscle group(measured as explosive strength).

Handgrip strength

What is handgrip strength?

Handgrip strength refers to the maximal isometric force thatcan be mainly generated by the hand and forehand musclesinvolved in the handgrip performance.

Why is handgrip strength important in the young population?

The handgrip strength test is a simple and economical test thatgives practical information on muscle, nerve, bone or jointdisorders. In adults, handgrip strength has been proposed as apossible predictor of mortality and the expectancy of beingable to live independently (Metter et al. 2002). Results fromthe AVENA study revealed a negative association between

handgrip strength and total cholesterol/HDL cholesterollipoprotein-related risk factors (Ortega et al. 2004).

Handgrip strength test methodology in young people

The handgrip strength test is a widely used test in experi-mental and epidemiological studies. The measure of handgripstrength is influenced by several factors, including age,gender, different angle of shoulder, elbow, forearm, and wrist(Richards et al. 1996), posture (Watanabe et al. 2005) andgrip span (Ruiz-Ruiz et al. 2002). Another important factoraffecting handgrip strength is hand size (Ruiz-Ruiz et al.2002; Ruiz et al. in press). The handgrip test was measuredin ∼3,000 Spanish adolescents in the framework of theAVENA study. Detailed test methodology and referencevalues have been properly described elsewhere (Ortega et al.2005; Ruiz et al. in press). Briefly, subjects performed thetest in a standard bipedal position and with the arm incomplete extension without touching any part of the bodywith the dynamometer except the hand being measured.

We made an attempt to find the optimal grip span thatresulted in maximum handgrip strength and that increasedreliable and reproducible handgrip strength in adult popula-tion (Ruiz-Ruiz et al. 2002). Recently, we have shown astandard procedure to evaluate the maximum handgripstrength in adolescents (Ruiz et al. in press). The results ofour study suggest that there is an optimal grip span to whichthe dynamometer should be adjusted when measuringhandgrip strength in young subjects. For males, the optimalgrip span can be derived from the equation y=x/7.2+3.1 cmand for females y=x/4+1.1 cm, where y is optimal grip spanand x is hand size measured from the tip of the thumb to thetip of the little finger with the hand open widely. Theseequations may improve the validity and accuracy of resultsand may guide clinicians and researchers in selecting theoptimal grip span on the hand dynamometer when measuringhandgrip strength in young, healthy subjects.

“Curl-up”

What is the “curl-up” test?

The curl-up test assesses trunk strength, i.e. abdominalmuscular endurance. Muscular endurance is the ability of amuscle group to execute repeated contractions over time orto maintain a maximal voluntary contraction for a pro-longed period of time.

Why is performing curl-up important in the youngpopulation?

The strength of abdominal muscles has been shown to havea significant association with lower back pain in adults

J Public Health (2006) 14:269–277 273

(Nourbakhsh and Arab 2002). Improvements in abdominalmuscle strength have been shown to not only reduce lowback pain but also to prevent injury recurrence in athletes(Trainor and Trainor 2004), and young adults (Arokoski etal. 2001). Low back pain is a common and costly complaintin society. Its multifactorial aetiology is not well under-stood, but it is assumed to involve biomechanical loadingof the spine and psychosocial influences (Keyserling 2000).Also, overweight (Leboeuf-Yde 2000), smoking (Goldberget al. 2000) and lack of physical exercise (Hildebrandt et al.2000) may contribute to low back pain. To prospectivelyevaluate the influence of low abdominal strength in youngpeople with the likelihood of developing low back painlater in life would be of special interest from a public healthperspective.

“Curl-up” test methodology in young people

The cadence-based curl-up test is the recommended test forabdominal strength/endurance testing in the FITNESS-GRAM battery (Cooper Institute for Aerobics Research1999). The curl-up test is a modification of the traditionalsit-up test included in the EUROFIT battery (Committee ofExperts on Sports Research EUROFIT 1993). The differ-ences between the former and the full sit up are armplacement, leg position and range of motion of movement.Moreover, the reduced action of the psoas iliac muscle inthe curl-up test may prevent back pain when performing thetest. The use of a cadence (25 reps per minute) with the curlup also seems to eliminate many concerns about theballistic nature of 30-s (or 1-min) all-out speed tests. Inaddition, the use of a cadence allows students to focus ontheir own performance and avoid competitive speeding up.

Standing broad jump and Bosco jumps

What are standing broad jumps and Bosco jumps?

The standing broad jump assesses lower-limb explosivestrength. Explosive strength is the ability to carry out amaximal, dynamic contraction of a muscle or muscle group.It is the maximum rate of working of a muscle or musclegroup. In the HELENA study, a more detailed assessmentof muscle performance of the lower limbs has beenproposed. Different jump tests will be measured accordingto the Bosco protocol. The Bosco jump protocol includes,among other things, the following type of jumps: squatjump, countermovement jump and Abalakov jump. Perfor-mance in squat jump indicates explosive strength of thelower limbs; the countermovement jump assesses explosivestrength plus the use of elastic energy; the Abalakov jumpassesses explosive strength, plus the use of elastic energy,plus the coordinative capacity using trunk and upper limbs.

These are usually performed by young subjects (Vicente-Rodriguez et al. 2003, 2004a).

Why is standing broad jump important in the youngpopulation?

Jump performance together with speed has been shown tobe highly strongly correlated with mean hip and lumbarbone mass accretion (Vicente-Rodriguez et al. 2003,2004a). Results from the AVENA study revealed a negativeassociation between standing broad jump and total choles-terol in overweight/obese male adolescents (Fig. 2) (Ortegaet al. 2004).

From a public health perspective, these observations areof greater interest mainly because the standing broad jumptest is an easy and feasible test to be used in schools; infact, it is preformed as a part of the curriculum in manyEuropean countries.

Standing broad jump test methodology in young people

The standing broad jump test is a simple and cost- andtime-effective test and is part of the EUROFIT battery(Committee of Experts on Sports Research EUROFIT1993). The subject is instructed to push off vigorouslyand jump as far as possible trying to land with both feettogether. The score is the distance from the take-off line tothe point where the back of the heel nearest to the take-offline lands on the mat. Reference values of a population

Fig. 2 Associations between standing broad jump and total choles-terol in overweight/obese Spanish adolescents. Overweight/obesecategory was categorised following the International Obesity TaskForce (IOTF)-proposed gender- and age-adjusted body mass index(BMI) cutoff points

274 J Public Health (2006) 14:269–277

sample of Spanish adolescents participating in the AVENAstudy and a detailed description of the test can be foundelsewhere (Ortega et al. 2005).

Bosco jump protocol

A more detailed and accurate information about muscleperformance of the lower limbs can be obtained by use of theBosco system (ERGOJUMP Plus, BOSCO SYSTEM, Byo-medic, S.C.P., Barcelona, Spain). Briefly, the Ergojump Boscosystem measures flight time during the vertical jump. Thisapparatus consists of a digital timer (±0.001 s) connected by acable to two infrared bars. The timer is triggered by the feet ofthe subject at the moment of release from the platform andstops at the moment of contact coming down. As mentioned,the Bosco jump protocol includes three types of jumps (squat,countermovement and Abalakov) measuring different musclecharacteristics. Briefly, the tests are performed as follows: in thesquat jump, the subject performs a vertical jump starting from ahalf-squat position, with trunk straight and both hands on hipsand without doing a previous countermovement; the counter-movement jump is similar to the previous one, but the legs areextended in the start position, and a flexion–extension of thelegs must be performed as fast as possible; finally, theAbalakov jump is a natural vertical jump. The results fromthese tests allow the calculation of relevant muscle-strength-related indexes, such as the elasticity index [measures elasticenergy = ({counter movement jump − squat jump}/countermovement jump)x100] and the upper limbs coordination index[({Abalakov − countermovement jump})/Abalakov)×100].Moreover, the software allows estimation of the percentage offast-twitch fibres (Bosco et al. 1983).

“Bent-arm hang”

What is the “bent-arm hang” test?

The bent-arm hang assess upper-limb endurance strength.This test evaluates the ability to maintain a maximalvoluntary contraction (hanging from a bar) for a prolongedperiod of time, i.e. assesses mainly the arm, shoulder anddorsal muscular endurance. It is proposed as a marker offunctional strength.

Why is performing “bent-arm” hang importantin the young population?

Results from the AVENA study suggest that the bent-armhang test is positively associated with HDL cholesterol andwith total cholesterol to HDL cholesterol ratio (Fig. 1), aswell as with body fat, expressed as the sum of six skinfolds,and/or percentage of body fat estimated by the Slaughterequation (FB Ortega, JR Ruiz, MJ Castillo, A Gutierrez,

unpublished data, 2006). Deforche et al. (2003) showed thatobese subjects had significantly lower performances onbent-arm hang and other weight-bearing tasks comparedwith their non-obese counterparts; however, the obese hadbetter results in handgrip strength test. These results supportfindings from the AVENA study. The bent-arm hang testhas been shown to be a significant explanatory factor foradult health-related fitness in Finnish female pupils studiedfrom 9 to 21 years of age (Mikkelsson et al. 2006).

“Bent-arm hang” test methodology in young people

The bent-arm hang test (also called flexed arm hang) is one ofthe recommended tests for upper-limb endurance strength inboth the FITNESSGRAM battery (Cooper Institute forAerobics Research 1999) and the EUROFIT battery (Com-mittee of Experts on Sports Research EUROFIT 1993).Reference values of a population sample of Spanish adoles-cents participating in the AVENA study and detailed method-ology of the test can be found elsewhere (Ortega et al. 2005).

Speed of movement/agility

This is the ability of a specific muscle or muscle group beable to move as quickly as possible over a distance.

Shuttle run (4×10-m)

What is the shuttle run (4×10-m)?

The shuttle run test (4×10-m) assesses the subjects’ speed ofmovement, agility and coordination in an integrated fashion.

Why is performing shuttle run (4×10-m) importantin the young population?

Preliminary results from the AVENA study have shown astrong independent relationship between speed (assessed bymeans of 4×10-m shuttle-run test) and bone mineral contentin both male and female adolescents, regardless of the stageof maturation (G Vicente-Rodriguez, MI Mesana, LAMoreno, JR Ruiz, FB Ortega, M Bueno, unpublished data,2006). Recently, it has been shown that some physical-fitness-related variables, specifically those related withspeed and dynamic strength, had a high predictive valuefor both bone mineral content and density and also for theaccumulation of bone mass during early puberty (Vicente-Rodriguez et al. 2003, 2004a,b).

Shuttle run test (4x10-m) methodology in young people

The shuttle run (4×10-m) test is a modification of theshuttle run (10×5-m) test included in the EUROFIT battery

J Public Health (2006) 14:269–277 275

(Committee of Experts on Sports Research EUROFIT1993). The present test also includes four sponges that arecarried one by one to the different lines. The subjects runback and forth four times along a 10-m track at the highestspeed possible. At the end of each track section, thesubjects deposit or pick up a sponge from a line on thefloor. Therefore, it allows measurement not only speed ofdisplacement but also agility and coordination. Validationstudies have been done in our university, and results willsoon be published. Detailed methodology and referencevalues from the AVENA study have been reportedelsewhere (Ortega et al. 2005).

Concluding comment

Results and experiences obtained from pan-Europeanresearch suggest that physical fitness is a key healthmarker in children and adolescents. The fitness tests tobe included in the assessment of health-related fitnessin the HELENA study seem to give relevant informa-tion regarding the health status of the young people andare feasible and objective. Validation studies of mosttests are already done (Ruiz et al. in press) and othersare under the validation process. Future health informa-tion systems should include monitoring of health-relatedfitness among adults as well as among young individuals,and results and experiences from recent and ongoingresearch projects on young people across Europe, such asAVENA, EYHS and HELENA studies, should be takenadvantage of. Some of these experiences have beensummarised in this review. Relevant methodology seemsto be available. Development of efficient systems forlarge-scale collection of health-related fitness data andtransfer of data to centrally located databases will be thenext step. The working party “Lifestyle” within theHealth Information Strand of the Public HealthProgramme 2003–2008 of the EC (DG SANCO) hasdeveloped an implementation and dissemination strategyto put into operation and ensure rapid transfer of data andexperiences to the units within the commission, nationalhealth authorities and other stakeholders involved in thedevelopment and implementation of health informationsystems.

Acknowledgements The AVENA study was funded by the SpanishMinistry of Health, FEDER-FSE funds FIS no. 00/0015, CSD grants05/UPB32/0, 109/UPB31/03 and 13/UPB20/04, the Spanish Ministryof Education (AP2003-2128; AP-2004-2745), and scholarships fromPanrico S.A., Madaus S.A. and Procter and Gamble S.A. The Swedishpart of the EYHS was supported by grants from the StockholmCounty Council (MS), and the Estonian part of EYHS was supportedby a grant from the Estonian Science Foundation No. 3277 and 5209and by the Estonian Centre of Behavioural and Health Sciences. TheHELENA study takes place with the financial support of the European

Community Sixth RTD Framework Programme (Contract FOOD-CT-2005-007034). The content of this article reflects only the authors’views, and the European Community is not liable for any use that maybe made of the information contained therein. The authors of thepresent paper are responsible partners of the physical activity andfitness assessment in the HELENA Study.

Conflict of interest statement No benefits in any form have beenreceived or will be received from a commercial party related directlyor indirectly to the subject of this article. None of the authors had anyconflict of interest.

References

Arokoski JP, Valta T, Airaksinen O, Kankaanpaa M (2001) Back andabdominal muscle function during stabilization exercises. ArchPhys Med Rehabil 82:1089–1098

Australian Sports Commission (1999) 20m shuttle run test: aprogressive shuttle run test for measuring aerobic fitness.Belconnen (ACT): Australian Coaching Council

Baquet G, Twisk JW, Kemper HC, Van Praagh E, Berthoin S (2006)Longitudinal follow-up of fitness during childhood: interactionwith physical activity. Am J Hum Biol 18:51–58

Bosco C, Luhtanen P, Komi PV (1983) A simple method formeasurement of mechanical power in jumping. Eur J ApplPhysiol 50:273–282

Bouchard C (1986) Genetics of aerobic power and capacity. In Sportsand Human Genetics. Human Kinetics, Champaign, IL

Brewer J, Ramsbottom R, Williams C (1988) Multistage fitness test: aprogressive shuttle-run test for the prediction of maximumoxygen uptake. National Coaching Foundation, Leeds

Castillo Garzon MJ, Ortega Porcel FB, Ruiz Ruiz J (2005)[Improvement of physical fitness as anti-aging intervention].Med Clin 124:146–155

Committee of Experts on Sports Research EUROFIT (1993) Hand-book for the EUROFIT Tests of Physical Fitness. Council ofEurope, Strasburg, GE

Cooper Institute for Aerobics Research (1999) FITNESSGRAM testadministration manual. Human Kinetics, Champaign

Deforche B, Lefevre J, De Bourdeaudhuij I, Hills AP, Duquet W,Bouckaert J (2003) Physical fitness and physical activity in obeseand nonobese Flemish youth. Obes Res 11:434–441

Goldberg MS, Scott SC, Mayo NE (2000) A review of the associationbetween cigarette smoking and the development of nonspecificback pain and related outcomes. Spine 25:995–1014

Gonzalez-Gross M, Ruiz JR, Moreno LA, De Rufino-Rivas P,Garaulet M, Mesana MI, Gutierrez A; AVENA Group (2003)Body composition and physical performance of Spanish adoles-cents: the AVENA pilot study. Acta Diabetol 40:S299–S301

Hansen HS, Froberg K, Nielsen JR, Hyldebrandt N (1989) A newapproach to assessing maximal aerobic power in children: theOdense School Child Study. Eur J Appl Physiol Occup Physiol58:618–624

Hildebrandt VH, Bongers PM, Dul J, van Dijk FJ, Kemper HC (2000)The relationship between leisure time, physical activities andmusculoskeletal symptoms and disability in worker populations.Int Arch Occup Environ Health 73:507–518

Izaak SI, Panasiuk TV (2005) [The age- and sex specificity ofphysical development of schoolchildren]. Gig Sanit 5:61–64

Jonker JT, De Laet C, Franco OH, Peeters A, Mackenbach J,Nusselder WJ (2006) Physical activity and life expectancy withand without diabetes: life table analysis of the Framingham HeartStudy. Diabetes Care 29:38–43

276 J Public Health (2006) 14:269–277

Kemper HC, Twisk JW, van Mechelen W, Post GB, Roos JC, Lips P(2000) A fifteen-year longitudinal study in young adults on therelation of physical activity and fitness with the development ofthe bone mass: The Amsterdam growth and health longitudinalstudy. Bone 27:847–853

Keyserling WM (2000) Workplace risk factors and occupationalmusculoskeletal disorders: Part 1. A review of biomechanical andpsychophysical research on risk factors associated with low-backpain. Am Ind Hyg Assoc J 61:39–50

Kibler WB, Goldberg C, Chandler TJ (1991) Functional biomechan-ical deficits in running athletes with plantar fasciitis. Am J SportsMed 266:185–196

Klasson-Heggebo L, Andersen LB, Wennlof AH, Sardinha LB, HarroM, Froberg K, Anderssen SA (2006) Graded associationsbetween cardiorespiratory fitness, fatness, and blood pressure inchildren and adolescents. Br J Sports Med 40:25–29

Koutedakis Y, Bouziotas C (2003) National physical educationcurriculum: motor and cardiovascular health related fitness inGreek adolescents. Br J Sports Med 37:311–314

Leach RE, James S, Wasilewski S (1981) Achilles tendinitis. Am JSports Med 9:93–98

Leboeuf-Yde C (2000) Body weight and low back pain: a systematicliterature review of 56 journal articles reporting on 65 epidemi-ologic studies. Spine 25:226–237

Léger LA, Lambert A, Goulet A, Rowan C, Dinelle Y (1984)Capacity aerobic des Quebecois de 6 a 17 ans - test Navette de 20metres avec paliers de 1 minute. Can J Appl Sport Sci 9:64–69

Liu NYS, Plowman SA, Looney MA (1992) The reliability andvalidity of the 20-meter shuttle test in American students 12 to 15years old. Res Q Exerc Sport 63:360–365

McHugh M, Connolly D, Eston R, Kremenic IJ, Nicholas SJ,Gleim GW (1999) The role of passive stiffness in symptomsof exercise-induced muscle damage. Am J Sports Med27:594–599

Mesa JL, Ruiz JR, Ortega FB, Warnberg J, Gonzalez-Lamuno D,Moreno LA, Gutierrez A, Castillo MJ (2006a) Aerobic physicalfitness in relation to blood lipids and fasting glycaemia inadolescents: Influence of weight status. Nutr Metab CardiovascDis 16:285–293

Mesa JL, Ortega FB, Ruiz JR, Castillo MJ, Hurtig Wennlöf A,Gutiérrez A (2006b) The importance of cardiorespiratory fitnessfor healthy metabolic traits in children and adolescents. TheAVENA Study. J Public Health 14:178–180

Metter EJ, Talbot LA, Schrager M, Conwit R (2002) Skeletalmuscle strength as a predictor of all-cause mortality in healthymen. J Gerontol A Biol Sci Med Sci 57:B359–B365

Mikkelsson L, Kaprio J, Kautiainen H, Kujala U, Mikkelsson M,Nupponen H (2006) School fitness tests as predictors of adulthealth-related fitness. Am J Hum Biol 18:342–349

Mokdad AH, Marks JS, Stroup DF, Gerberding JL (2004) Actualcauses of death in the United States, 2000. JAMA 291:1238–1245

Monyeki MA, Koppes LL, Kemper HC, Monyeki KD, Toriola AL,Pienaar AE, Twisk JW (2005) Body composition and physicalfitness of undernourished South African rural primary schoolchildren. Eur J Clin Nutr 59:877–883

Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood JE(2002) Exercise capacity and mortality among men referred forexercise testing. N Engl J Med 346:793–801

Nourbakhsh MR, Arab AM (2002) Relationship between mechanicalfactors and incidence of low back pain. J Orthop Sports PhysTher 32:447–460

Ortega FB, Ruiz JR, CastilloMJ,Moreno LA,Gonzalez-GrossM,WarnbergJ, Gutierrez A; Grupo AVENA (2005) [Low level of physical fitness inSpanish adolescents. Relevance for future cardiovascular health(AVENA study)]. Rev Esp Cardiol 58:898–909

Ortega FB, Ruiz JR, Gutiérrez A, Moreno LA, Tresaco B, MartínezJA, González-Lamuño D, Wärnberg J, Castillo MJ, and theAVENA Study group (2004) Is physical fitness a good predictorof cardiovascular disease risk factors in normal-weight andoverweight or obese adolescents? The AVENA Study. [abstract]Int J Obes Relat Metab Disord 28:S120

Ortega FB, Tresaco B, Ruiz JR, Moreno LA, Martin-Matillas M,Warnberg J, Bueno M, Tercedor P, Gutierrez A, Castillo MJ(2006) Cardiorespiratory fitness and sedentary activity areassociated with adiposity in adolescents. Obesity (in press)

Pope R, Herbert R, Kirwan J, Grahan BJ (2000) A randomized trial ofpre-exercise stretching for prevention of lower limb injury. MedSci Sports Exer 32:271–277

Richards LG, Olson B, Palmiter-Thomas P (1996) How forearmposition affects grip strength. Am J Occup Ther 50:133–138

Riddoch CJ (1990) The Northern Ireland health and fitness survey -1989: the fitness, physical activity, attitudes and lifestyles ofNorthern Ireland post-primary schoolchildren. The Queen’sUniversity of Belfast, Belfast

Ruiz JR, Rizzo N, Wennlof A, Ortega FB, Harro M, Sjostrom M(2006a) Relations of total physical activity and intensity to fitnessand fatness in children; The European Youth Heart Study. Am JClin Nutr 84:299–303

Ruiz JR, Ortega FB, Meusel D, Harro M, Oja P, Sjöström M (2006b)Cardiorespiratory fitness is associated with features of metabolicrisk factors in children. Should cardiorespiratory fitness beassessed in a European health monitoring system? The EuropeanYouth Heart Study. J Public Health 14:94–102

Ruiz JR, España-Romero V, Ortega FB, Sjöström M, Castillo MJ,Gutiérrez A on behalf of the HELENA Study (in press) Hand-size influences optimal grip span in male and femaleadolescents. J Hand Surgery (Am)

Ruiz-Ruiz J, Mesa JLM, Gutiérrez A, Castillo MJ (2002) Hand sizeinfluences optimal grip span in women but not in men. J HandSurg 27A:897–901

Strong WB, Malina RM, Blimkie CJ, Daniels SR, Dishman RK, GutinB, Hergenroeder AC, Must A, Nixon PA, Pivarnik JM, RowlandT, Trost S, Trudeau F (2005) Evidence based physical activity forschool-age youth. J Pediatr 146:732–737

Trainor TJ, Trainor MA (2004) Etiology of low back pain in athletes.Curr Sports Med Rep 3:41–46

Vicente-Rodriguez G, Jimenez-Ramirez J, Ara I, Serrano-Sanchez JA,Dorado C, Calbet JA (2003) Enhanced bone mass and physicalfitness in prepubescent footballers. Bone 33:853–859

Vicente-Rodriguez G, Ara I, Perez-Gomez J, Serrano-Sanchez JA,Dorado C, Calbet JA (2004a) High femoral bone mineral densityaccretion in prepubertal soccer players. Med Sci Sports Exerc36:1789–1795

Vicente-Rodriguez G, Dorado C, Perez-Gomez J, Gonzalez-Henriquez JJ, Calbet JA (2004b) Enhanced bone mass andphysical fitness in young female handball players. Bone35:1208–1215

Wärnberg J (2006) Inflammatory status in adolescents; the impact ofhealth determinants such as overweight and fitness. Thesis fordoctoral degree, Karolinska University Press, Stockholm

Watanabe T, Owashi K, Kanauchi Y, Mura N, Takahara M, Ogino T(2005) The short-term reliability of grip strength measurementand the effects of posture and grip span. J Hand Surg 30:603–609

Witvrouw E, Lysens R, Bellemans J, Cambier D, Vanderstraeten G(2000) Intrinsic risk factors for the development of anterior kneepain in an athletic population: a two year prospective study. Am JSports Med 28:480–489

Witvrouw E, Bellemans J, Lysens R, Danneels L, Cambier D (2001)Intrinsic risk factors for the development of patellar tendinitis inthe athletic population. Am J Sports Med 29:190–195

J Public Health (2006) 14:269–277 277

CARDIORESPIRATORY FITNESS IS ASSOCIATED WITH

FEATURES OF METABOLIC RISK FACTORS IN CHILDREN;

SHOULD CARDIORESPIRATORY FITNESS BE ASSESSED IN

A EUROPEAN HEALTH MONITORING SYSTEM?

THE EUROPEAN YOUTH HEART STUDY

Jonatan R. Ruiz1,2, Francisco B. Ortega1,2, Dirk Meusel3, Maarike

Harro4, Pekka Oja1, Michael Sjöström2

J Public Health 2006; 14: 94-102

1Unit for Preventive Nutrition, Department of Biosciences and Nutrition at

NOVUM, Karolinska Institutet, Huddinge, Sweden, 2Department of Physiology,

School of Medicine, University of Granada, Granada, Spain, 3Research

Association Public Health, Institute of Clinical Pharmacology, Medical Faculty,

Dresden University of Technology, Germany, 4National Institute for Health

Development, Tallinn, Estonia; and Estonian Centre of Behavioural and Health

Sciences.

IV

J Public Health (2006) 14: 94–102DOI 10.1007/s10389-006-0026-8

ORIGINAL ARTICLE

Jonatan R. Ruiz . Francisco B. Ortega . Dirk Meusel .Maarike Harro . Pekka Oja . Michael Sjöström

Cardiorespiratory fitness is associated with features of metabolic

risk factors in children. Should cardiorespiratory fitness

be assessed in a European health monitoring system?

The European Youth Heart Study

Received: 19 December 2005 / Accepted: 11 January 2006 / Published online: 16 March 2006# Springer-Verlag 2006

Abstract The question as to whether fitness should beassessed in a European health monitoring system, perhapsfrom the early stages of life onwards, remains to beanswered. We aimed to examine the associations betweencardiorespiratory fitness and metabolic risk factors inchildren. A total of 873 healthy children from Sweden andEstonia aged 9–10 years (444 girls and 429 boys) wererandomly selected. A maximal ergometer bike test wasused to estimate cardiorespiratory fitness. Additionalcardiovascular risk factors were assessed. Significantdifferences among cardiorespiratory fitness quartiles forthe sum of five skinfolds, insulin resistance, triglycerides,and total cholesterol (TC) and high-density lipoproteincholesterol (HDLc) ratio were shown in girls whereas inboys, the sum of five skinfolds and insulin resistance weresignificantly different. The lowest sum of five skinfolds

and insulin resistance was shown in the highest cardio-respiratory fitness quartile in girls and boys, and the lowestvalues of triglyceride and TC/HDLc values in the highestcardiorespiratory fitness quartile was observed only ingirls. Cardiorespiratory fitness was negatively associatedwith a clustering of metabolic risk factors in girls and boys.The results add supportive evidence to the body ofknowledge suggesting that cardiorespiratory fitness inchildren is an important health marker and thus should beconsidered to be included in a pan-European healthmonitoring system.

Keywords Cardiorespiratory fitness . Children .Metabolic syndrome . Cardiovascular diseases

Introduction

Low cardiorespiratory fitness seems to be an importanthealth problem (Lee et al. 1999; Carnethon et al. 2003;Mora et al. 2003; Myers et al. 2002). It has been recentlyshown that low cardiorespiratory fitness is a strong andindependent predictor of incident metabolic syndrome (i.e.hypertension, dyslipidemia, impaired glycemic control andobesity) in men and women (LaMonte et al. 2005), whichcould be one of the mechanism of overall cardiovasculardisease. Moreover, cardiorespiratory fitness seems toprevent premature mortality regardless of body-weightstatus or the presence of metabolic syndrome in adult men(Katzmarzyk et al. 2004, 2005).

High cardiorespiratory fitness during childhood andadolescence has been associated not only with healthiercardiovascular profile during these years but also later inlife (Twisk et al. 2002). However, the association betweencardiorespiratory fitness and cardiovascular risk factors inchildren is uncertain, probably because of low researchpriority. Furthermore, most children are asymptomatic forcardiovascular disease. Cardiorespiratory fitness has beensuggested to be included in the European Health Monitor-ing System for the adult population (Sjöström et al. 2005),

J. R. Ruiz . F. B. Ortega . P. Oja . M. Sjöström (*)Unit for Preventive Nutrition, Biosciences at NOVUM,Karolinska Institutet,Huddinge, Swedene-mail: [email protected].: +46-8-608-9140Fax: +46-8-608-3350

J. R. Ruiz . F. B. OrtegaDepartment of Physiology, School of Medicine,University of Granada,Granada, Spain

D. MeuselResearch Association Public Health Saxonyand Saxony-Anhalt, Technische Universität Dresden,Dresden, Germany

M. HarroNational Institute for Health Development and Estonian Centreof Behavioural and Health Sciences,Tallinn, Estonia

M. SjöströmUnit for Preventive Nutrition,Department of Biosciences and Nutrition, Karolinska Institutet,14157 Novum, Huddinge, Sweden

but the question as to whether fitness should be assessed inEuropean health monitoring systems from the early stages oflife remains to be answered. Understanding the associationbetween a low cardiorespiratory fitness and cardiovascular-disease-related outcomes in children would support thequestion as to whether cardiorespiratory fitness might ormight not be proposed as a health marker at these ages.Therefore, the aim of the present report was to examine theassociations of cardiorespiratory fitness to health-relatedvariables in a wide cohort of children aged 9–10 years and torelate the findings with corresponding results from recentcross-sectional and prospective cohort studies.

Research design and methods

The present cross-sectional study involved 873 childrenaged 9–10 years (444 girls, 429 boys). The subjectscomprised Estonian and Swedish children who were part ofthe European Youth Heart Study (EYHS) (Poortvliet et al.2003). The pooling of data was assumed to be possiblebecause of the use of common protocols in both countries(Poortvliet et al. 2003; Wennlof et al. 2003). Study design,selection criteria and sample calculations have beenreported elsewhere (Riddoch et al. 2005).

In Estonia, the city of Tartu and its surrounding rural areawas the geographical sampling area. In Sweden, sevenmunicipalities in the Stockholm area and one in Örebrowere chosen for data collection. The local ethicalcommittees approved the study (University of Tartu no.49/30-1997, University Hospital no. 474/98 Huddinge, andÖrebro City Council no. 690/98). The study procedureswere explained verbally and in written text to all parentsand children. One parent or legal guardian provided writteninformed consent, and all children gave verbal consent.

Data collection

Physical examination

Height and weight were measured by standardizedprocedures. Body mass index was calculated as weight/height squared (kg/m2). Skinfold thicknesses were mea-sured with a Harpenden caliper at the biceps, triceps,subscapular, suprailiac and triceps surae areas on the leftside of the body. These measures have been shown tohighly correlate with dual-energy X-ray absorptiometry-measured body fat percentage in children of similar ages(Gutin et al. 1996). All measurements were taken twice andin rotation, and the mean was calculated. If the differencebetween the two measurements differed by >2 mm, a thirdmeasurement was taken, and the two closest measurementswere averaged. The sum of five skinfold thicknesses wasused as an indicator of body fat.

Blood pressure

The systolic and diastolic blood pressures were measuredwith an automatic oscillometric method (Dinamap modelXL Critikron, Inc., Tampa, Florida.). The equipment hasbeen validated in children (Park and Menard 1987). Anappropriate cuff size was chosen according to themanufacturer’s recommendation after checking the armcircumference. The subject was in a sitting, relaxedposition, and recordings were made every second minutefor 10 min with the aim of obtaining a set of systolicrecordings not varying by more than 5 mmHg. The meanvalue of the last three recordings was used as the restingsystolic and diastolic blood pressure in millimeters ofmercury (mmHg).

Blood samples

With the subject in the supine position, blood samples weretaken by venipuncture after an overnight fast, usingvacuum tubes (Vacuette, Greiner Lab Technologies Inc).The fasting state was verbally confirmed by the subjectbefore blood sampling. Blood was centrifuged for 10 minat 2,000 g, serum was separated within 30–60 min, and thesamples were stored at −80°C. Serum concentrations oftriglycerides, total cholesterol (TC), high-density lipopro-tein cholesterol (HDLc), and glucose were measured on anOlympus AU600 autoanalyser (Olympus DiagnosticaGmbH, Hamburg, Germany). The insulin for the Estoniansubjects was analyzed with an enzyme immunoassay(DAKO Diagnostics Ltd., Ely, England). All analyseswere performed at Bristol Royal Infirmary, UK, with theexception of insulin for the Swedish subjects, which wasperformed at Huddinge University Hospital, Sweden(Elecsys, Roche Diagnostics GmbH, Mannheim, Ger-many). A more detailed description of the blood analysishas been reported elsewhere (Wennlof et al. 2005). Insulinresistance was estimated from fasting glucose and insulinaccording to the homeostasis model assessment (HOMA)(Matthews et al. 1985), and the ratio TC/HDLc was alsocalculated.

Cardiorespiratory fitness test

Cardiorespiratory fitness was determined by a maximumcycle-ergometer test, as described elsewhere (Hansen et al.1989). Briefly, the workload was preprogrammed on acomputerized cycle ergometer (Monark 829E Ergomedic,Vansbro, Sweden) to increase every third minute untilexhaustion. Heart rate was registered continuously bytelemetry (Polar Sport Tester, Kempele, Finland). Criteriafor exhaustion were a heart rate ≥185 beats per minute,failure to maintain a pedaling frequency of at least 30revolutions per minute, and a subjective judgment by theobserver that the child could no longer keep up, even aftervocal encouragement. The power output was calculatedas =W1+(W2 · t/180), where W1 is a work rate at fully

95

completed stage, W2 is the work rate increment at finalincomplete stage, and t is time in second at final incompletestage. The “Hansen formula” for calculated maximumoxygen consumption (VO2max) in ml/min was = 12 xcalculated power output + 5 x body weight in kg (Hansen etal. 1989). Cardiorespiratory fitness was expressed asVO2max per kilogram of body mass.

Metabolic risk score

The metabolic risk score was computed from the followingsix variables: insulin, glucose, HDLc, triglycerides, the

sum of five skindfolds, and blood pressure (systolic anddiastolic blood pressure). Each of these variables wasstandardized as follow: standardized value = (value −mean)/SD. The HDLc standardized value was multipliedby −1 to indicate higher cardiovascular risk with increasingvalue. The standardized values of systolic and diastolicblood pressure were averaged. The metabolic risk scorewas compiled by the sum of the six standardized scoresdivided by six. The resulting risk score is a continuousvariable with a mean of zero by definition, with lowerscores denominating a more favorable profile.

Table 1 Baseline characteristics of 873 children (444 girls, 429 boys)

Girls Boys

Mean 95% CI Mean 95% CI

Age (years) 9.54 9.50–9.58 9.58 9.54–9.63Height (m) 1.28 1.37–1.39 1.38 1.38–1.39Weight (kg) 32.03 31.45–32.60 32.11 31.60–32–63Body mass index (kg/m2) 16.73 16.52–16.94 16.76 16.57–16.94Sum of five skinfolds (mm) 44.65 42.96–46.35 37.67 34.32–37.01Insulin (mU/L) 6.44 6.11–6.77 5.47 5.17–5.77Glucose (mg/dl) 87.98 87.39–88.58 91.26 90.67–91.85Insulin resistance 1.42 1.34–1.49 1.25 1.17–1.32High density lipoprotein cholesterol (mg/dl) 55.22 54.16–56.27 57.61 56.54–58.69Total cholesterol (mg/dl) 176.76 173.70–179.42 170.41 167.87–167.87Triglycerides (mg/dl) 68.83 66.47–71.19 60.35 57.80–62.89Systolic blood pressure (mmHg) 101.92 101.10–102.74 103.08 102.21–103.95Diastolic blood pressure (mmHg) 60.65 60.00–61.29 60.10 59.41–60.79Metabolic risk score 0.03 −0.01–0.08 −0.03 −0.08–0.01Cardiorespiratory fitness (ml/kg/min) 37.16 36.69–37.63 43.06 42.48–43.63

25

30

35

40

45

50

55

60

65

Sum

of

5 sk

info

lds

(mm

)

Girls

Boys

1 2 3 4

Cardiorespiratory fitness(quartiles)

*

†‡

Fig. 1 Associations between sum of five skinfolds and cardio-respiratory fitness quartiles in girls and boys. Data shown as meanand 95% confidence interval (CI). Girls in the first quartile (*) had ahigher sum of five skinfolds than in superior quartiles (P<0.001),and girls in the second quartile (†) had a higher sum of five skinfoldsthan in the fourth quartile (P=0.004). Boys in the first quartile (‡)had a higher sum of five skinfolds than in superior quartiles.(P=0.007)

1.0

1.2

1.4

1.6

1.8

2.0

Insu

lin r

esis

tanc

e (H

OM

A)

Girls

Boys

1 2 3 4


* †

‡

Fig. 2 Associations between insulin resistance estimated from thehomeostasis model assessment (HOMA) equation and cardiore-spiratory fitness quartiles in girls and boys. Data shown as mean and95% confidence interval (CI). Girls in the first quartile (*) had ahigher HOMA than in the fourth quartile (P<0.001), and girls in thesecond quartile (†) had a higher sum of five skinfolds than in thefourth quartile (P<0.001). Boys in the first quartile (‡) had a higherHOMA than in the fourth quartile (P=0.007)

96

Statistical analysis

All variables were checked for normality of distributionbefore the analysis, and appropriate transformations wereapplied when necessary. Sum of five skinfolds, triglycer-ides, low-density lipoprotein cholesterol (LDLc), TC, andTC/HDLc were logarithmically transformed, and HOMAwas transformed by taking it by the power of (1/3).Differences between metabolic syndrome individual vari-ables and cardiorespiratory fitness quartiles, and metabolicsyndrome risk score and cardiorespiratory fitness quartileswere assessed by analysis of variance (ANOVA). Differ-ences of metabolic syndrome individual variables amongcardiorespiratory fitness quartiles were assessed by Tukey’stest. All analyses were performed using the StatisticalPackage for Social Sciences (SPSS, version 13.0 forWINDOWS; SPSS Inc, Chicago, IL,USA), and the level ofsignificance was set at P<0.05.

Results

The descriptive characteristics of the study sample areshown in Table 1. All subjects in this study were within thenormal healthy ranges for all studied variables. TheANOVA showed significant differences among cardio-respiratory fitness quartiles for sum of five skinfolds,insulin resistance, triglycerides and TC/HDLc in girlswhereas in boys, only sum of five skinfolds and insulinresistance were significantly different. Significant differ-ences among cardiorespiratory fitness quartiles were alsoobserved for metabolic risk score in girls and boys.

The Tukey’s test showed that the sum of five skinfoldswas significantly higher in the first cardiorespiratory fitnessquartile compared with the second, third and fourthcardiorespiratory fitness quartiles in girls and boys(Fig. 1). Moreover, sum of five skinfolds was significantlyhigher in the second cardiorespiratory fitness quartilecompared with the fourth cardiorespiratory fitness quartilein girls. In boys, the sum of five skinfolds was significantlylower in the fourth cardiorespiratory fitness quartilecompared with the first, second and third cardiorespiratoryfitness quartiles (Fig. 1).

Insulin resistance was significantly higher in the firstcardiorespiratory fitness quartile compared with the fourthcardiorespiratory fitness quartile in both girls and boys.Moreover, insulin resistance was significantly higher in thesecond cardiorespiratory fitness quartile compared with thefourth cardiorespiratory fitness quartile in girls (Fig. 2).

Triglyceride values were significantly higher in the firstcardiorespiratory fitness quartile compared with the fourthcardiorespiratory fitness quartile in girls (Fig. 3). The ratioof TC/HDLc was significantly higher in the first cardio-respiratory fitness quartile compared with the second andfourth cardiorespiratory fitness quartiles in girls (Fig. 4).

50.0

55.0

60.0

65.0

70.0

75.0

80.0

TG

(m

g/dl

)Girls

Boys

1 2 3 4


*

Fig. 3 Associations between triglycerides (TG) and cardiorespira-tory fitness quartiles in girls and boys. Data shown as mean and 95%confidence interval (CI). Girls in the first quartile (*) had a higherTG values than in the fourth quartile (P<0.001)

2.8

3.0

3.2

3.4

3.6

3.8

TC

/HD

Lc

Girls

Boys*

1 2 3 4


Fig. 4 Associations between total cholesterol (TC) and high-densitylipoprotein cholesterol (HDLc) ratio and cardiorespiratory fitnessquartiles in girls and boys. Data shown as mean and 95% confidenceinterval (CI). Girls in the first quartile (*) had a higher TC/HDLcratio than in the second and fourth quartiles (P<0.001)

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

Met

abol

ic r

isk

Z s

core

Girls

Boys

1 2 3 4


*†

‡

Fig. 5 Associations between metabolic risk score and cardiore-spiratory fitness quartiles in girls and boys. Data shown as mean and95% confidence interval (CI). Girls in the first quartile (*) had ahigher risk score than in the second, third and fourth quartiles(P<0.007), and girls in the second quartile (‡) had a higher risk scorethan in the fourth quartile (P<0.02). Boys in the first quartile (†) hada higher risk score than in the second, third and fourth quartiles(P=0.007)

97

Tab

le2

Sum

maryof

recent

cross-sectionalstudiesexam

iningtheassociations

betweencardiorespiratoryfitnessandhealth-related

variablesin

child

renandadolescents

Author/study

Subjects

Age

(years)

Outcomevaria

bles

Results

Boreham

etal.2

001

Boys

=251

12TC

,HDLc

,systolic

BP,d

iasto

licBP

,sum

offour

skinfolds

Boys

andgirls

TheNorthernIre

land

YoungHearts

Project

Girls=258

CRFwas

inverselyassociated

with

TC,T

C/HDLc

,and

systo

licBP

,but

was

notindependent

offatness

Boys

=252

15

Girls=254

Nielse

nandAndersen2003

Boys

=5,464

15–2

0Bloodpressure

Boys

Girls=8,093

TheORof

hyperte

nsionin

thelowestC

RFquintilecomparedto

thehighestC

RFquintile

was

1.3(P<0

.04),a

ftera

djustfor

ageandBM

IGirls

TheORof

hyperte

nsionin

thelowestC

RFquintilecomparedwith

thehighestC

RFquintile

was

1.5(P<0

.001),aftera

djustfor

ageandBM

I

Brageet

al.2

004

Boys

=279

8–10

TG,H

DLc

,sum

offour

skinfolds,insulin

,glucose,systolic

BP,a

nddiastolic

BPBo

ysandgirls

European

YouthHeartStudy(D

enmark)

Girls=380

Metabolic

syndromeZscore

CRFwas

inverselyassociated

with

insulin

,TG,systolic

BP,a

ndskinfold

thicknesses(P≤0

.033)

CRFwas

inverselyassociated

with

metabolic

syndromeZscore(P≤0

.031)

CRFwas

positivelyassociated

with

HDLc

(P=0

.002)

Gutin

etal.2

004

Boys

=116

14–1

8Insulin

,glucose

Boys

andgirls

Girls=166

CRFwas

inverselyassociated

with

insulin

concentra

tions,a

ndtheadverseim

pact

oflow

CRF

was

greaterinboys

than

ingirls

Reed

etal.2

005

Boys

=55

9–11

BP,%

BF,a

rteria

lcom

pliance(la

rgeandsm

all)

Boys

andgirls

Girls=44

CRFaccountedfor3

7%of

thevaria

ncein

largeartery

compliance.

HighestCR

Fquartile

hadgreaterc

ompliancethan

child

renin

thetwolowestC

RFquartiles

byas

muchas

34%

Eisenm

annet

al.2

005a

Boys

=416

9–18

TG,T

C,HDLc

,LDLc

,glucose,B

P,BM

IBo

ysandgirls

Quebecfamily

study

Girls=345

CRFandBM

Ishowed

anindependenta

ssociatio

nwith

cardiovascular

riskfactors

Gutin

etal.2

005

Boys

=187

14–1

8TG

,TC,

HDLc

,LDLc

,LDLsz,

Lp(a),BM

I,WC,

%BF

Boys

andgirls

Girls=211

HigherC

RFandlower

fatnesswereassociated

with

favorablelip

idprofile.F

ormostv

ariables,

fatnesswas

slightly

greaterthantheinflu

ence

ofCR

F

TCtotalcholesterol,HDLchigh-density

lipoprotein

cholesterol,LDLclow-density

lipoprotein

cholesterol,LDLsz

LDLparticlesize,T

Gtriglycerides,BPbloodpressure

Lp(a)

lipoprotein

(a),apoapolipoprotein,CRFcardiorespiratoryfitness,BMIbody

massindex,

ORodds

ratio

,%BFpercentage

ofbody

fat

98

Tab

le3

Sum

maryof

recent

prospectivecohortstudiesexam

iningtheassociations

betweencardiorespiratoryfitnessandhealth-related

variablesin

child

renandadolescents

Author/study

Yearsof

follo

w-up

Subjects

Age

(years)

Outcomevaria

bles

Results

Boreham

etal.2

002

10Bo

ys=229

12and15

to22.5

TC,H

CLc,

systo

licBP

,diasto

licBP

,sum

offour

skinfolds

Boys

TheNorthernIre

land

YoungHearts

Project

1989/90–1992/93–1997/99

Girls=230

CRFchangesweremodestly

associated

with

TC,H

DLc

,and

systo

licBP

(P>0

.5)

Girls

CRFchangesweremodestly

associated

with

TC,H

DLc

,and

skinfold

thicknesses

(P>0

.17),a

ndsig

nific

antly

associated

with

diastolic

BP(P=0

.03)

Hasselstrøm

etal.2

002

8Bo

ys=133

15–19to

23–2

7TG

,HDLc

,systolic

BP,d

iasto

licBP

,%BF

Boys

Danish

youthandsportsstu

dy1983

–1991

Girls=132

Risk

score

CRFchangesbetween1983

–1991wereinverselycorre

latedwith

thechangesin

TC,T

G,H

DLc

/TC(P<0

.01)

Boys

=45

Girls

Girls=57

CRFchangesbetween1983

–1991wereinverselycorre

latedwith

thechanges

inTG

,systolic

BP,%

BF,a

ndriskscore(P<0

.05)

Janz

etal.2

002

5Bo

ys=63

10.5

to15

TC,H

CLc,

LDLc

,sum

of6skinfolds,WC

Boys

andGirls

TheMuscatin

eStudy

Girls=62

CRFchangesbetweenyear

1to

5wereinverselycorre

latedwith

thechanges

insum

ofsix

skinfoldsandWC

(P<0

.05)

Twisk

etal.2

002

20Bo

ys=132

13to

32TC

,HDLc

,systolic

BP,d

iasto

licBP

,sum

offour

skinfolds,W/H

Boys

andgirls

TheAmste

rdam

Growth

andHealth

Longitu

dinalS

tudy

Girls=145

Therelatio

nshipbetweenCR

Fdurin

gtheadolescencewas

inverselyassociated

with

TC,sum

offour

skinfolds,andW/H

(P<0

.05)

Ferre

iraet

al.2

003

24with

9repeated

measurements

Boys

=75

13.1

to36

Carotid

intim

a–media

thicknessandstiffn

ess

ofthecarotid

,fem

oral,a

ndbrachial

arterie

sBo

ysandgirls

Girls=79

CRFchangeswerenota

ssociatedwith

carotid

intim

a–media

thickness

CRFchangeswereassociated

with

largeartery

stiffn

ess(P<0

.05)

Andersenet

al.2

004

8Bo

ys=133

16–19to

24–2

7TG

,TC/HDLc

,systolic

BP,%

BFBo

ysandgirls

Eighty

ears

follo

w–upin

theDanish

YouthandSp

ortS

tudy

Girls=172

CRFwas

associated

with

cardiovascular

diseaseriskfactors.Th

eprobability

for“

acase”at

thefirstexam

inationto

be“a

case”at

thesecond

was

6.0

Boreham

etal.2

004

8Bo

ys=251

12–15to

20–2

5Arte

rialstiffness

Boys

andgirls

TheNorthernIre

land

YoungHearts

Project

Girls=203

CRFwas

inverselyassociated

with

arteria

lstiffness

Eisenm

annet

al.2

005b

∼11

Boys

=36

15.9

to27.2

TG,T

C,HDLc

,glucose,systolic

BP,

diastolic

BP,B

MI,WC,

%BF

Boys

andgirls

TheAerobicsCe

nter

Longitu

dinalS

tudy

Girls=12

Adolescents’

CRFisrelatedonly

toadultB

MI,WCand%BF

(P<0

.05)

Ferre

iraet

al.2

005

23Bo

ys=175

13to

36Prevalence

ofthemetabolic

syndrome

Boys

andgirls

TheAmste

rdam

Growth

andHealth

Longitu

dinalS

tudy

Girls=189

CRFchangeswereinverselyassociated

with

prevalence

ofmetabolic

syndrome

TCtotalcholesterol,HDLchigh

density

lipoprotein

cholesterol,LDLclow

density

lipoprotein

cholesterol,TG

triglycerides,BPbloodpressure,Lp(a)

lipoprotein

(a),apoapolipoprotein,

CRFcardiorespiratoryfitness,BMIbody

massindex,

WCwaistcircum

ference,

W/H

waistto

hipratio

,%BFpercentage

ofbody

fat

99

Metabolic risk score was significantly higher in the firstcardiorespiratory fitness quartile than in the second, thirdand fourth cardiorespiratory fitness quartiles in girls andboys (Fig. 5). Significant differences were also foundbetween metabolic risk score in the second and fourthcardiorespiratory fitness quartiles in girls (Fig. 5).

Discussion

The association between cardiorespiratory fitness andfeatures of metabolic syndrome was investigated in apopulation sample of Swedish and Estonian children aged9–10 years. Cardiorespiratory fitness was negativelyassociated with a clustering of metabolic risk factors ingirls and boys, and the lowest values of sum of fiveskinfolds, insulin resistance, triglyceride and TC/HDLcwere in the highest cardiorespiratory fitness quartile.

Theses results may suggests that cardiorespiratoryfitness should be proposed as a health marker in children.In fact, it is biologically plausible that a high cardio-respiratory fitness provides more health protection thanlow cardiorespiratory fitness, even in healthy children aswell as it has been found in adults (Balady 2002; Myers etal. 2002; Carnethon et al. 2003; Gulati et al. 2003; Kurl etal. 2003; Mora et al. 2003; Church et al. 2005; Katzmarzyket al. 2004, 2005; LaMonte el al. 2005). Risk-factor levelsare lower in children than in adults, but similar patternshave been seen in children. Previous cross-sectional studiesin children have shown significant associations betweencardiorespiratory fitness and plasma lipids and betweencardiorespiratory fitness and clustering of metabolic syn-drome risk factors (Table 2). In our study, triglyceride andTC/HDLc values differed among cardiorespiratory fitnessquartiles (Fig. 3). Moreover, negative associations betweenincreased cardiorespiratory fitness and clustering of met-abolic syndrome risk factors in both girls and boys havebeen shown here (Fig. 5). Cardiorespiratory fitness hasrecently been associated with arterial compliance inchildren aged 9–11 years, which may support the conceptthat fitness may exert a protective effect on the cardiovas-cular system (Reed et al. 2005).

Associations between cardiorespiratory fitness and car-diovascular risk factors have also been found in adoles-cents (Table 2). Gutin et al. (2004) found inverseassociations between cardiorespiratory fitness and insulinconcentrations. Furthermore, inverse associations betweencardiorespiratory fitness and the likelihood of havinghypertension were shown in 15- to 20-year-old subjects(Nielsen and Andersen 2003). In the present study, insulinresistance was significantly lower in the fourth cardio-respiratory fitness quartile compared with the firstcardiorespiratory fitness quartile in both girls and boys(Fig. 2). However, no differences were found in systolic ordiastolic blood pressure among cardiorespiratory fitnessquartiles (data not shown).

A summary of recent prospective cohort studiesexamining the associations between cardiorespiratoryfitness and health-related variables in children andadolescents is shown in Table 3. A number of longitudinalstudies have suggested that a low cardiorespiratory fitnessduring childhood and adolescence is associated with latercardiovascular risk factors, such as hyperlipidemia, hyper-tension and obesity (Boreham et al. 2001, 2002;Hasselstrøm et al. 2002; Janz et al. 2002; Twisk et al.2002; Ferreira et al. 2005). In an 8-year follow-up study,fitness during adolescence was not associated to riskfactors of cardiovascular disease in adulthood, but changesin fitness from adolescence to adulthood were related torisk in adulthood. Moreover, subjects who decreased theirfitness levels also changed to a worse risk factor profile(Hasselstrøm et al. 2002). Changes in cardiorespiratoryfitness from adolescence to adulthood were also inverselyand significantly associated with large arterial stiffness (amajor risk factor for cardiovascular disease) (Ferreira et al.2003; Boreham et al. 2004). Taken together, these resultsseem to support the existence of a strong associationbetween cardiorespiratory fitness and health-related out-comes in the young population, which may suggest theimportance of including cardiorespiratory fitness tests inthe monitoring system.

The test used to calculate cardiorespiratory fitness inthis study was objectively and accurately measured, and ithas been previously validated in children of the same age(Riddoch et al. 2005). However, laboratory tests presentsome disadvantages, as necessity of sophisticated instru-ments, qualified technicians and cost and time constraints,and it may cause problems for the subjects to go to thelaboratory, etc. Therefore, in some circumstances, fieldtests may be a better option because a large number ofsubjects can be tested at the same time, as the tests aresimple, safe and often the only feasible methods.

The cross-sectional nature of this study limits the abilityto determine any causality in the results. We also do notknow if an extrapolation of the association may be madefor overweight and obese children or those with subclinicalmanifestations of cardiovascular pathologies. Neverthe-less, with regular reports of increasing childhood obesityand related disease prevalence world wide, the results ofthis study are noteworthy. The ideal study to answer thequestion as to whether high levels of cardiorespiratoryfitness during childhood lower the risk of developingcardiovascular diseases later in life is a randomizedcontrolled trial with a lifetime follow-up, in which a largenumber of children is assigned to either an active or asedentary life style.

In conclusion, the present study shows negativeassociations between cardiorespiratory fitness and featuresof metabolic syndrome in children aged 9–10 years. Theresults suggest that cardiorespiratory fitness in children, ashas been shown in adults, is potentially an important healthmarker and should be considered to be included in a pan-European health monitoring system.

100

Acknowledgements The Swedish part of the study was supportedby grants from the Stockholm County Council (MS), and theEstonian part was supported by a grant from the Estonian ScienceFoundation No. 3277 and 5209, and by Estonian Centre ofBehavioural and Health Sciences. JRR and FBO were supportedby a grant from Ministerio de Educación y Ciencia de España(AP2003-2128, AP2004-2745) and CSD (109/UPB31/03,13/UPB20/04).

References

Andersen LB, Hasselstrøm H, Grønfeldt V, Hansen SE, Froberg K(2004) The relationship between physical fitness and clusteredrisk, and tracking of clustered risk from adolescence to youngadulthood: eight years follow-up in the Danish Youth and SportStudy. Int J Behav Nutr Phys Act 1:6

Balady GJ (2002) Survival of the fittest-more evidence. N Engl JMed 346:852–854

Boreham C, Twisk J, Murray L, Savage M, Strain JJ, Cran G (2001)Fitness, fatness, and coronary heart disease risk in adolescents:the Northern Ireland Young Hearts Project. Med Sci SportsExerc 33:270–274

Boreham C, Twisk J, Neville C, Savage M, Murray L, Gallagher A(2002) Associations between physical fitness and activitypatterns during adolescence and cardiovascular risk factors inyoung adulthood: The Northern Ireland Young Hearts Project.Int J Sports Med 23:22S–26S

Boreham CA, Ferrera I, Twisk JW Gallagher AM, Savage MJ,Murray LJ (2004) Cardiorespiratory fitness, physical activity,and arterial stiffness: The Northern Ireland Young HeartsProject. Hypertension 44:721–726

Brage S, Wedderkopp N, Ekelund U, Franks PW, Wareham NJ,Andersen LB, Froberg K (2004) European Youth Heart Study(EYHS): Features of the metabolic syndrome are associatedwith objectively measured physical activity and fitness inDanish children: the European Youth Heart Study (EYHS).Diabetes Care 27:2141–2148

Carnethon MR, Gidding SS, Nehgme R, Sidney S, Jacobs DR, LiuK (2003) Cardiorespiratory fitness in young adulthood and thedevelopment of cardiovascular disease risk factors. JAMA290:3092–3100

Church TS, LaMonte MJ, Barlow CE, Blair SN (2005) Cardio-respiratory fitness and body mass index as predictors ofcardiovascular disease mortality among men with diabetes.Arch Intern Med 10:2114–21220

Eisenmann JC, Katzmarzyk PT, Perusse L, Tremblay A, Despres JP,Bouchard C (2005a) Aerobic fitness, body mass index, andCVD risk factors among adolescents: the Quebec family study.Int J Obes Relat Metab Disord 29:1077–1083

Eisenmann JC, Wickel EE, Welk GJ, Blair SN (2005b) Relationshipbetween adolescent fitness and fatness and cardiovasculardisease risk factors in adulthood: the Aerobics Center Longi-tudinal Study (ACLS). Am Heart J 149:46–53

Ferreira I, Twisk JW, Stehouwer CD, van Mechelen W, Kemper HC.(2003) Longitudinal changes in VO2max: associations withcarotid IMT and arterial stiffness. Med Sci Sports Exerc35:1670–1678

Ferreira I, Twisk JW, Stehouver CD, van Mechelen W, Kemper HC(2005) The metabolic syndrome, cardiopulmonary fitness, andsubcutaneous trunk fat as independent determinants of arterialstiffness: the Amsterdam growth and health longitudinal study.Arch Intern Med 25:875–882

Gulati M, Pandey DK, Arnsdorf MF, Lauderdale DS, Thisted RA,Wicklund RH, Al-Hani AJ, Black HR (2003) Exercise capacityand the risk of death in women: the St. James women take heartproject. Circulation 108:1554–1559

Gutin B, Litaker M, Islam S, Manos T, Smith C, Treiber F (1996)Body-composition measurement in 9-11-y-old children bydual-energy X-ray absorptiometry, skinfold thickness measure-ments, and bioimpedance analysis. Am J Clin Nutr 63:287–292

Gutin B, Yin Z, Humphries MC, Bassali R, Le NA, Daniels S,Barbeau P (2004) Relations of fatness and fitness to fastinginsulin in black and white adolescents. J Pediatr 145:737–743

Gutin B, Yin Z, Humphries MC, Bassali R, Le NA, Daniels S,Barbeau P (2005) Relations of body fatness and cardiovascularfitness to lipid profile in black and white adolescents. PediatrRes 58:78–82

Hansen HS, Froberg K, Nielsen JR, Hyldebrandt N (1989) A newapproach to assessing maximal aerobic power in children: theodense school child study. Eur J Appl Physiol Occup Physiol58:618–624

Hasselstrøm H, Hansen SE, Froberg K, Andersen LB (2002)Physical fitness and physical activity during adolescence aspredictors of cardiovascular disease risk in young adulthood.Danish youth and sports study. An eight-year follow-up study.Int J Sports Med 23:27S–31S

Janz KF, Dawson JD, Mahoney LT (2002) Increases in physicalfitness during childhood improve cardiovascular healthduring adolescence: The muscatine study. Int J Sports Med23:15S–21S

Katzmarzyk PT, Church TS, Blair SN (2004) Cardiorespiratoryfitness attenuates the effects of the metabolic syndrome on allcause and cardiovascular disease mortality in men. Arch InternMed 164:1092–1097

Katzmarzyk PT, Church TS, Janssen I, Ross R, Blair SN (2005)Metabolic syndrome, obesity, and mortality: impact ofcardiorespiratory fitness. Diabetes Care 28:391–397

Kurl S, Laukkanen JA, Rauramaa R, Lakka TA, Sivenius J, SalonenJT (2003) Cardiorespiratory fitness and the risk for stroke inmen. Arch Intern Med 163:1682–1688

LaMonte MJ, Barlow CE, Jurca R, Kampert JB, Church TS, BlairSN (2005) Cardiorespiratory fitness is inversely associated withthe incidence of metabolic syndrome: a prospective study ofmen and women. Circulation 26:505–512

Lee CD, Blair SN, Jackson AS (1999) Cardiorespiratory fitness,body composition, and all-cause and cardiovascular diseasemortality in men. Am J Clin Nutr 69:373–380

Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF,Turner RC (1985) Homeostasis model assessment: insulinresistance and beta-cell function from fasting plasma glucoseand insulin concentrations in man. Diabetologia 28:412–419

Mora S, Redberg RF, Cui Y, Whiteman MK, Flaws JA, Sharrett AR,Blumenthal R (2003) Ability of exercise testing to predictcardiovascular and all-cause death in asymptomatic women: a20-year follow-up of the lipid research clinics prevalence study.JAMA 290:1600–1607

Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood JE(2002) Exercise capacity and mortality among men referred forexercise testing. N Engl J Med 346:793–801

Nielsen GA, Andersen LB (2003) The association between highblood pressure, physical fitness, and body mass index inadolescents. Prev Med 36:229–234

Park MK, Menard SM (1987) Accuracy of blood pressuremeasurement by the Dinamap monitor in infants and children.Pediatrics 79:907–914

Poortvliet E, Yngve A, Ekelund U, Hurtig-Wennlöf A, Nilsson A,Hagströmer M, Sjöström M (2003) The European Youth HeartSurvey (EYHS). An international study that addresses themulti-dimensional issues of CVD risk factors. Forum Nutr56:254–256

Reed KA, Warburton DE, Lewanczuk RZ, Haykowski MJ, ScottJM, Whitney CL, McGavock JM, McKay HA (2005) Arterialcompliance in young children: the role of aerobic fitness. Eur JCardiovasc Prev Rehabil 12:492–497

101

Riddoch C, Edwards D, Page A, Froberg K, Anderssen SA,Wedderkopp N, Brage S, Cooper AR, Sardinha LB, Harro M,Klasson-Heggebø L, van Mechelen W, Boreham C, Ekelund U,Bo Andersen L, and the European Youth Heart Study team(2005) Cardiovascular disease risk factors in children: Ratio-nale, aims, study, design, and validation of methods. J Phys ActHealth 2:115–129

Twisk JW, Kemper HC, van Mechelen W (2002) The relationshipbetween physical fitness and physical activity during adoles-cence and cardiovascular disease risk factors at adult age. TheAmsterdam growth and health longitudinal study. Int J SportsMed 23:S8–S14

Sjöström M, Poortvliet E, Nelson M (2005) Monitoring publichealth nutrition in Europe: nutritional indicators and determi-nants of health status. J Public Health1 3:74–83

Wennlöf AH, Yngve A, Sjöström M (2003) Sampling procedure,participation rates and representativeness in the Swedish part ofthe European Youth Heart Study (EYHS). Public Health Nutr6:291–299

Wennlöf AH, Yngve A, Nilsson TK, Sjöström M (2005) Serumlipids, glucose and insulin levels in healthy schoolchildren aged9 and 15 years from Central Sweden: reference values inrelation to biological, social and lifestyle factors. Scand J ClinLab Invest 65:65–76

102

CARDIOVASCULAR FITNESS IS NEGATIVELY ASSOCIATED

WITH HOMOCYSTEINE LEVELS IN FEMALE ADOLESCENTS

Jonatan R. Ruiz1,2, Ricardo Sola1,3, Marcela Gonzalez-Gross4,

Francisco B. Ortega1,2, German Vicente-Rodirguez5, Miguel

García-Fuentes6, Ángel Gutiérrez1, Michael Sjöström2, Kalus

Pietrzik7, Manuel J. Castillo1

Arch Pediatr Adolesc 2007; 161: 166-171

1Department of Physiology, School of Medicine, University of Granada,


Nutrition at NOVUM, Karolinska Institutet, Huddinge, Sweden, 3Unit of

Hematology, University Hospital San Cecilio, 4School of Sport Sciences,

Universidad Politecnica de Madrid, 5E. U. Health Sciences, University of

Zaragoza, 6Department of Pediatrics, University of Cantabria, Santander,

Spain, 7Institut fuer Ernaehrungswissenschaft, Abt. Pathophysiologie der

Ernährung, Rheinische Friedrich-Wilhelms Universität, Bonn, Germany.

V

INFLAMMATORY PROTEINS ARE ASSOCIATED WITH

MUSCLE STRENGTH IN ADOLESCENTS

THE AVENA STUDY

Jonatan R. Ruiz1,2, Francisco B. Ortega1,2, Julia Wärnberg2,3, Luis

A. Moreno4, Juan J. Carrero5, Marcela González-Gross6,

Ascensión Marcos3, Ángel Gutiérrez1, Michael Sjöström2

Submitted

1Department of Physiology, School of Medicine, University of Granada,


Nutrition at NOVUM, Karolinska Institutet, Huddinge, Sweden, 3Immunonutrition Research Group, Department of Metabolism and Nutrition,

Consejo Superior de Investigaciones Científicas, Madrid, Spain, 4E. U. Health

Sciences, University of Zaragoza, 5Division of Renal Medicine and Baxter

Novum, Department of Clinical Science, Karolinska Institutet, Huddinge,

Sweden, 6School of Sport Sciences, Universidad Politecnica de Madrid.

VI

SUBMITTED

Inflammatory Proteins are Associated with Muscle Strength in Adolescents;

The AVENA Study

Jonatan R Ruiz, Francisco B Ortega, Julia Warnberg, Luis A Moreno, Juan J

Carrero, Marcela Gonzalez-Gross, Ascension Marcos, Angel Gutierrez, Michael

Sjöström

From the Department of Physiology, School of Medicine, University of Granada,

Granada, Spain (JRR, FBO, AG), the Unit for Preventive Nutrition, Biosciences at

NOVUM, Karolinska Institutet, Huddinge, Sweden (JRR, FBO, JW, MS), the

Immunonutrition Research Group, Department of Metabolism and Nutrition,

Consejo Superior de Investigaciones Científicas, Madrid, Spain (JW, AM), the E. U.

Health Sciences, University of Zaragoza (LAM), the Division of Renal Medicine and

Baxter Novum, Department of Clinical Science, Karolinska Institutet, Huddinge,

Sweden (JJC), the School of Sport Sciences, Universidad Politecnica de Madrid

(MGC).

The AVENA study was funded by the Spanish Ministry of Health, FEDER-FSE

funds FIS nº 00/0015, CSD grants 05/UPB32/0, 109/UPB31/03 and 13/UPB20/04,

the Spanish Ministry of Education (AP2003-2128; AP-2004-2745), and scholarships

from Panrico S.A., Madaus S.A. and Procter and Gamble S.A. JJC is supported by

an ERA-EDTA postdoctoral grant.

Address correspondence to Jonatan R. Ruiz. Department of Physiology, School of

Medicine, University of Granada, 18071, Granada, Spain. Tel: +34 958 243 540,

Fax: + 34 958 249 015, E-mail: [email protected]

Running head: Muscle strength in adolescents.

ABSTRACT

Background: Low-grade inflammation seems to be negatively associated with

cardiorespiratory fitness in overweight and non-overweight young person and adults.

Whether low-grade inflammation is associated with muscle strength in adolescents is

unknown.

Objective: The aim of this study was to examine the associations between

inflammatory proteins and muscle strength, and to determine whether this

association varies between overweight and non-overweight adolescents.

Design: C-reactive protein, complement factors C3 and C4, ceruloplasmin and

transthyretin were measured in 416 Spanish adolescents (230 boys and 186 girls)

aged 13 to 18.5 y. Muscle strength score was computed as the mean of the handgrip

and standing broad jump standardized values, and cardiorespiratory fitness was

measured by the 20 m shuttle run test. A muscle strength score was computed as the

mean of the handgrip and standing broad jump standardized values. The adolescents

were categorized as overweight and non-overweight according to body mass index.

Results: The analysis of covariance showed that C-reactive protein, C4 and

ceruloplasmin were negatively associated with muscle strength. C-reactive protein

and transthyretin were negatively associated with muscle strength in overweight

adolescents after adjusting for sex, age, pubertal status, socioeconomic status,

cardiorespiratory fitness and body fat.

Conclusions: Low-grade inflammation is negatively associated with muscle strength

in adolescents. The fact that some inflammatory proteins were associated with

muscle strength in overweight adolescents after adjusting for body fat indicate that

muscle mass may be involved in this mechanism. Intervention studies examining the

impact of strength training on inflammatory markers in adolescents are warranted.

Key Words: Inflammation, physical fitness, exercise, pediatrics, obesity.

INTRODUCTION

Low-grade inflammation seems to play a role in the development of cardiovascular

disease from on early age (1,2). It is negatively associated with cardiorespiratory

fitness and positively associated with body fat in young persons and adults (3-7).

Recent findings show a higher prevalence of having high C-reactive protein levels in

Spanish overweight unfit adolescents compared with their overweight fit

counterparts (8). Inflammatory proteins have also been negatively associated with

muscle strength in adults (9-11).

The role of muscle strength in the performance of exercise and activities of daily

living, as well as in preventing disease has become increasingly recognized (12,13).

Resistance exercise training increased muscle strength, and it is currently prescribed

by the major health organizations for improving health and fitness (14,15).

Cardiovascular disease risk factors have also been associated with aerobic exercise

and cardiorespiratory fitness not only in adults but also young persons (15-17).

Whether low-grade inflammation is associated with muscle strength in adolescents is

unknown. Therefore, the aim of the present study was to examine the associations

between inflammatory proteins and muscle strength in adolescents, and to determine

whether these associations vary in overweight and non-overweight adolescents.

SUBJECTS AND METHODS

Subjects

The study participants were adolescents aged 13 to 18.5 y from the AVENA study

(Alimentación y Valoración del Estado Nutricional de los Adolescentes Españoles

[Food and Assessment of the Nutritional Status of Spanish Adolescents]), which was

designed to assess the health and nutritional status of adolescents. The AVENA

study design and sampling procedure have been reported in detail elsewhere (18-10).

Data were collected from 2000 to 2002 in five Spanish cities, including Granada,

Madrid, Murcia, Santander and Zaragoza. After exclusion of nine adolescents with

concentrations of C-reactive protein >10mg/L, the present article includes 416

adolescents (230 boys and 186 girls) whom had a complete set of inflammatory

proteins, muscle strength and cardiorespiratory fitness measurements. A

comprehensive verbal description of the nature and purpose of the study was given to

both the adolescents and their teachers. Written consent to participate was requested

from parents and adolescents, and all adolescents gave verbal assent. Adolescents

with a personal history of cardiovascular disease, taking medication at the time of the

study, or were pregnant, were excluded after completion of the field work. The study

protocol was performed in accordance with the ethical standards establised in the

1961 Declaration of Helsinki (as revised in Hong-Kong in 1989, and in Edinburgh in

2000), and was approved by the Review Committee for Research Involving Human

Subjects of the Hospital Universitario Marqués de Valdecilla (Santander, Spain).

The parents completed a questionnaire, which addressed the adolescents’ previous

and current health status. Socioeconomic status was also assessed in the questionnaire,

and was defined by the educational achievement and occupation of the father.

According to this information, and following the recommendation of the Spanish

Society for Epidemiology, the adolescents were classified into five socioeconomic

categories: low, medium-low, medium, medium-high and high socioeconomic status.

Physical Examination

Anthropometric measurements were obtained as described elsewhere (19,21,22).

Body mass index (BMI) was calculated as weight/height squared (kg/m2). Skinfold

thickness was measured at the biceps, triceps, subscapular, suprailiac, thigh and calf

on the left side of the body to the nearest 0.2 mm using a Holtain skinfold caliper.

The sum of the six skinfold thicknesses was used as an indicator of total body fat

(22).

BMI categories (non-overweight and overweight including obesity) were computed

according the proposed gender- and age-adjusted BMI cut-off points derived from

adult values associated with health risk (23). Overweight prevalence and

anthropometric body fat composition values in the complete AVENA study have

been described by Moreno et al (21,22).

Identification of pubertal status (I-V) was assessed according to Tanner and

Whitehouse (24). The standard staging of pubertal maturity describes breast and

pubic hair development in girls and genital and pubic hair development in boys.

Blood Sampling

After overnight fasting, blood samples were collected between 8:00 and 9:30 a.m. by

venipuncture. Highly sensitive C-reactive protein, complement factors C3 and C4,

and ceruloplasmin were measured by immunoturbidimetry (AU2700 biochemistry

analyzer; Olympus, Rungis, France). Transthyretin was measured by

immunoturbidimetry (Roche/Hitachi 912). Quality control of the assays was assured

by the Regional Health Authority. A detailed description of the blood analysis has

been already reported (20,25).

Muscle Strength

Upper body strength was assessed by handgrip strength test, and lower body strength

was assessed by the standing broad jump test. The handgrip strength test was

performed on both hands with a hand dynamometer (Takei T.K.K. 5101 Grip-D;

Takey, Tokyo, Japan) standing, and with the arm completely extended. The

dynamometer was in contact with the hand being measured only, and no other part of

the body. The standing broad jump was performed in an indoor rubber floored

gymnasium. The subjects were instructed to push off vigorously and jump as far

forward as possible, trying to land on both feet. The score was the distance from the

take-off line to the point where the back of the heel closest to the take-off line lands

on the floor.

A muscle strength score was computed by combining the standardized values of

handgrip strength and standing broad jump. Each of these variables was standardized

as follows: standardized value = (value - mean)/ standard deviation. The

standardized values of the handgrip strength obtained with the right and the left hand

were averaged. The muscle strength score was calculated as the mean of the two

standardized scores (handgrip strength and standing broad jump). The score was

calculated separately for boys and girls and for each age group (13, 14, 15, 16, 17-

18.5 y).

Cardiorespiratory Fitness

Cardiorespiratory fitness was assessed by the 20 m shuttle run test as previously

described (26). In brief, participants were required to run between two lines 20 m

apart. The initial speed was 8.5 km/hr, which was increased by 0.5 km/hr per minute

(one minute equal to one stage). The test was finished when the subject failed to

reach the end lines concurrent with the audio signals on two consecutive occasions.

Otherwise, the test ended when the subject stopped because of fatigue.

Cardiorespiratory fitness was considered as the number of stages completed.

The adolescents were instructed to abstain from strenuous exercise for the 48 hours

preceding the fitness tests. The tests are part of the EUROFIT test battery, and have

been validated and standardized by the Council of Europe (27). Detailed

methodology and reference values of fitness tests performed in the AVENA Study

have been reported by Ortega et al (28).

Data Analysis

The data are presented as means ± SDs, unless otherwise indicated. All the residuals

showed a satisfactory pattern after skinfold thickness, C-reactive protein, C3, C4,

ceruloplasmin and transthyretin were normalized by natural logarithm

transformation. Gender differences were analyzed by one-way analysis of variance

(ANOVA), and adjusted for mass significance as described by Holm (29). Nominal

data (overweight/non-overweight, pubertal status and socioeconomic status) were

analyzed using Chi-square tests. Partial correlations were used to examine bivariate

relations between cardiorespiratory fitness and muscle strength after controlling for

sex.

The association between inflammatory proteins and muscle strength was tested by

one-way analysis of covariance (ANCOVA). Muscle strength was recoded into

tertiles to be entered into the models. All the analyses were adjusted for age, pubertal

status, weight, height, socioeconomic status and cardiorespiratory fitness. Since no

interaction effects between sex and muscle strength was found, all the analyses were

performed for boys and girls together.

The association between inflammatory proteins and BMI was tested by one-way

ANCOVA after adjustment for age, pubertal status, socioeconomic status and

cardiorespiratory fitness. BMI was entered into the models as overweight and non-

overweight.

To determine whether the association between inflammatory proteins and muscle

strength varies between BMI categories (overweight and non-overweight), the

analyses were performed by one-way ANCOVA separately in overweight and non-

overweight adolescents after adjusting for sex, age, pubertal status, socioeconomic

status and cardiorespiratory fitness. Because BMI does not discriminate between

muscle and fat mass, all the analyses were repeated with an additional adjustment

made for skinfold thickness (as an indicator of total body fat). The analyses were

performed using the Statistical Package for Social Sciences (SPSS, v. 14.0 for

WINDOWS; SPSS Inc, Chicago), and the level of significance was set to 0.05.

RESULTS

Data Completeness and Baseline Characteristics

All subjects (n = 416) had complete data for all variables measured, with the

exception of except for pubertal status and socioeconomic status data, which were

not available in 37 (9%) and 83 (20%) adolescents, respectively. The descriptive

characteristics of the study sample are shown in Table 1. Adolescent boys had higher

values of ceruloplasmin than adolescent girls, as well as higher values of

cardiorespiratory fitness, handgrip strength and standing broad jump.

Cardiorespiratory fitness was significantly associated with both handgrip strength

and standing broad jump (r = 0.148, P < 0.01 and r = 0.746, P < 0.001, respectively)

as well as with muscle strength score (r = 0.339, P < 0.001) after controlling for sex.

Inflammatory Proteins and Muscle Strength

The results of the associations between inflammatory proteins and muscle strength

are shown in Table 2. C-reactive protein and ceruloplasmin were negatively

associated with muscle strength after adjusting for sex, age, pubertal status, weight,

height, socioeconomic status, and cardiorespiratory fitness. C4 was not statistically

significantly associated with muscle strength (P for trend = 0.071). The results were

similar when additional adjustment was made for body fat (expressed as sum of six

skinfold thicknesses), except for ceruloplasmin, which was not significantly

associated with muscle strength.

Inflammatory Proteins and BMI

The associations between inflammatory proteins and BMI are shown in Table 3. C-

reactive protein, C3 and C4 were significantly associated with BMI after adjusting

for sex, age, pubertal status, socioeconomic status, and cardiorespiratory fitness. No

inflammatory protein was associated with BMI once the analysis was additionally

adjusted for total body fat, except C4, which remained significantly associated with

BMI.

Inflammatory Proteins and Muscle Strength by BMI Categories

The associations between inflammatory proteins and muscle strength by BMI

categories are shown in Figure 1. C-reactive protein and transthyretin were

negatively associated with muscle strength in overweight adolescents after adjusting

for sex, age, pubertal status, socioeconomic status, and cardiorespiratory fitness.

Ceruloplasmin was not statistically significantly associated with muscle strength in

overweight adolescents (P for trend = 0.058). C3 and C4 were not significantly

associated with muscle strength either in either overweight or in non-overweight

adolescents (P for trend > 0.1). The associations between C-reactive protein,

transthyretin and muscle strength remained significant (P for trend = 0.05 and 0.013,

respectively) after the analysis were additionally adjusted for total body fat.

Ceruloplasmin was not significantly (P for trend > 0.1) associated with muscle

strength once total body fat was entered in the model.

DISCUSSION

The primary findings of this study show that 1) C-reactive protein, C4 and

ceruloplasmin are negatively associated with muscle strength in adolescence; 2) C-

reactive protein and transthyretin are associated with muscle strength in overweight

adolescents after adjusting for different confounders including cardiorespiratory

fitness and body fat. Moreover, it also shows that the increased low-grade

inflammation found in overweight adolescents is mediated by body fat, which

confirms previous findings (3,4,25). To the best of our knowledge, this is the first

population based study showing that low-grade inflammation is associated with

muscle strength in adolescents.

Inflammatory Proteins and Muscle Strength

The association between inflammatory proteins and muscle strength has been

examined in a few studies in adults. Two cross-sectional studies have shown a

negative association of C-reactive protein, interleukin-6 and tumor-necrosis factor-

with muscle strength (9,10). Additionally, one prospective study found that higher

levels of interleukin-6 and C-reactive protein were associated with loss of muscle

strength in older persons (11). Features of the metabolic syndrome have also been

negatively associated with muscle strength in adult men (30). Findings from a

prospective study in adult men suggested that muscle strength may exert additive

protection against the incident of metabolic syndrome beyond that attributed to

cardiorespiratory fitness, and that overweight men may obtain more benefits than

non-overweight men (31).

The health benefits of cardiorespiratory fitness among young persons and adults are

well established (7,15-17,32). A number of studies on young persons suggest that

inflammatory proteins are negatively associated with cardiorespiratory fitness (3-6).

Similar findings have also been obtained in Spanish adolescents from the AVENA

study (8). The results of the current investigation suggest that the development of

muscle strength may confer additional benefits beyond those attributed to

cardiorespiratory fitness. Therefore, the results of the present study add supportive

evidence to the body of knowledge suggesting that physical fitness in young persons

is an important health marker.

High concentration of C-reactive protein is considered a major cardiovascular risk

factor (1,2,33). Increasing evidence supports the link between abnormal C3 and C4

concentrations and vascular disease (34). Body fat is known to promote a state of

low-grade inflammation (35), which lends credibility to the results obtained in the

present study. Furthermore, higher concentrations of inflammatory proteins have

been hypothesized to play a role in the functional decline of older persons (9,36,37).

The causal pathway leading from inflammation to disability has not been fully

explained, but it has been suggested that low-grade inflammation may cause a

decline of physical functioning through its catabolic effects on skeletal muscle (38).

Collectively, these mechanisms may give explanation to the observed association

between C-reactive protein and muscle strength in overweight adolescents.

Transthyretin, also known as pre-albumin, is a negative acute-phase protein that

declines in response to inflammation (39). Other factors such as starvation and

decreased skeletal muscle function are also known to affect transthyretin

concentrations (40). Transthyretin concentation has been shown to increase with

increasing protein and calorie intake and to decrease when protein intake is

inadequate (41). Therefore, the associations observed in our study between

transthyretin levels and muscle strength in overweight adolescents could be

explained by the putative association to muscular weakness, but also enhanced by the

state of increased low-grade inflammation seen in the overweight adolescents as

mentioned previously.

Inflammatory Proteins, Muscle Strength, BMI and Body Composition

It is noteworthy that the observed associations between C-reactive protein,

transthyretin, and muscle strength in overweight adolescents remained significant

after adjusting for body fat. This may indicate that other mechanisms beyond body

fat are involved in these associations. The key role of muscle mass in a number of

metabolic processes has been recently highlighted, as well as in the prevention of

many common pathologic conditions and chronic diseases (12,13). Measures of fat

free mass are not available in the adolescents of the AVENA study. However, its has

been recently shown that obese children are heavier not only due to an excess of fat

mass but also due to the higher levels of fat free mass (measured by dual X-ray

absorptiometry) compared with non-obese children (42).

Therefore, the theoretical increased muscle mass in overweight adolescents may

partially explain why those with high levels of muscle strength (third tertile) are

those with the lowest levels of C-reactive protein. Collectively, these findings

indicate that special efforts should be focused on sub-groups of adolescents at

increased risk of early cardiovascular disease, such as the overweight/obese. As a

first step, promotion of regular participation in strength activities may be of help,

since this mode of exercise may be easier and better tolerated for overweight/obese

youth than aerobic training (43). A limitation of weight-bearing activities at the start

of an intervention for overweight/obese adolescent is recommended, and a bigger

focus on non-weight-bearing activities and activities relying on muscle strength is

suggested (43,44). Interventions that are not tailored to the fitness level of obese

participants can be counterproductive, and may contribute to discouragement of

future participation in physical activity.

Cardiovascular Health and Muscle Strength

Strength training may have a number of beneficial effects for overweight individuals,

including increased muscle mass and decreased total and central fat mass (14,15). A

recent study has shown that a 16 week resistance training program (2 days/week)

significantly increased insulin sensitivity in overweight adolescent boys, independent

of changes in body composition, suggesting that mechanisms other than alterations

in body composition were operative for the enhanced insulin sensitivity (46).

Resistance exercise has been also successful in improving brachial artery endothelial

function in women (47), improving insulin sensitivity and fasting glycaemia, and

decreasing abdominal fat in both adult men and women with type 2 diabetes (48,49).

Nevertheless, further studies are required in order to show whether resistance

exercise can effectively attenuate the moderately increased resting levels of

inflammatory proteins as well as reduce the fat mass in overweight adolescents.

Increases in muscle strength may also influence positively the levels of

cardiorespiratory fitness, since both variables are significantly associated.

Study Limitation

The observations of the current study are limited by the cross-sectional design.

Prospective and intervention studies are required in order to draw more robust

conclusions on the determining effect of inflammatory proteins on muscle strength.

We used a single blood measurement of inflammation that may not accurately reflect

long-term inflammatory status. Although no subject with a known underlying cause

of infection was included, we can not be sure that elevated concentrations were not

due to the onset of an infection. However, to attempt to minimize the confounding

effect of an ongoing infection, adolescents with high concentrations of C-reactive

protein (>10mg/L) were not included in this study. The fitness tests included in the

present study have been shown to be reliable, as well as simple, safe and feasible,

especially in the schools setting, and are included in several fitness batteries to be

performed in school-based epidemiological studies (27,50).

Conclusions

The results presented in this study suggest that low-grade inflammation is negatively

associated with muscle strength in adolescents. The patterns of these associations

seem more relevant in overweight adolescents. The fact that some inflammatory

proteins were associated with muscle strength in overweight adolescents after

adjusting for body fat may indicate that muscle mass may be involved in this

mechanism. More studies are needed in order to elucidate the role of muscle mass

and muscle strength on low-grade inflammation. Intervention studies examining the

impact of strength training on muscle mass and inflammatory markers in adolescents

are warranted.

ACKNOWLEDGEMENTS

The authors thank Prof. Manuel J Castillo (Department of Physiology, University of

Granada, Spain) for highly valuable comments, and Prof. Olle Carlsson (Unit for

Preventive Nutrition, Department of Biosciences and Nutrition at NOVUM,

Karolinska Institutet, Sweden) for statistical assistance.

JRR conceived the hypothesis and conducted the statistical analyses for this

manuscript. JRR drafted the manuscript. FOP, LAM, JW, JJC, MGG, ACM, AG and

MS, contributed to the interpretation and discussion of the results. AM, MGG and

AGS contributed to the concept and design of the AVENA study. All the authors

critically revised the drafted manuscript. None of the authors had any conflict of

interest.

REFERENCES

1. Jarvisalo MJ, Harmoinen A, Hakanen M, et al. Elevated serum C-reactive protein

levels and early arterial changes in healthy children. Arterioscler Thromb Vasc Biol

2002; 22:1323-1328.

2. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J

Med 2005; 352:1685-1695.

3. Cook DG, Mendall MA, Whincup PH, et al. C-reactive protein concentration in

children: relationship to adiposity and other cardiovascular risk factors.

Atherosclerosis 2000; 149:139-150.

4. Nemet D, Wang P, Funahashi T, Matsuzawa Y, Tanaka S, Engelman L, Cooper

DM. Adipocytokines, body composition, and fitness in children. Pediatr Res 2003;

53:148-152.

5. Isasi CR, Deckelbaum RJ, Tracy RP, Starc TJ, Berglund L, Shea S. Physical fitness

and C-reactive protein level in children and young adults: the Columbia University

BioMarkers Study. Pediatrics 2003; 111:332-338.

6. Williams MJ, Milne BJ, Hancox RJ, Poulton R. C-reactive protein and

cardiorespiratory fitness in young adults. Eur J Cardiovasc Prev Rehabil 2005;

12:216-220.

7. LaMonte MJ, Durstine JL, Yanowitz FG, Lim T, DuBose KD, Davis P, Ainsworth

BE. Cardiorespiratory fitness and C-reactive protein among a tri-ethnic sample of

women. Circulation 2002; 106:403-406.

8. Wärnberg J, Ruiz JR, Sjöström M, et al. Association of fitness and fatness to low-

grade systemic inflammation in adolescents. The AVENA study. Med Sci Sports

Exer [abstract] 2006; 38:S8.

9. Taaffe DR, Harris TB, Ferrucci L, Rowe J, Seeman TE. Cross-sectional and

prospective relationships of interleukin-6 and C-reactive protein with physical

performance in elderly persons: MacArthur studies of successful aging. J Gerontol A

Biol Sci Med Sci 2000; 55: M709-715.

10. Visser M, Pahor M, Taaffe DR, et al. Relationship of interleukin-6 and tumor

necrosis factor-alpha with muscle mass and muscle strength in elderly men and

women: the Health ABC Study. J Gerontol A Biol Sci Med Sci 2002; 57:M326-332.

11. Schaap LA, Pluijm SM, Deeg DJ, Visser M. Inflammatory markers and loss of

muscle mass (sarcopenia) and strength. Am J Med 2006; 119:526.e9-17.

12. Stump CS, Henriksen EJ, Wei Y, Sowers JR. The metabolic syndrome: Role of

skeletal muscle metabolism. Ann Med 2006; 38:389-420.

13. Wolfe RR. The underappreciated role of muscle in health and disease. Am J Clin

Nutr 2006; 84:475-482.

14. Pollock ML, Franklin BA, Balady GJ, et al. AHA Science Advisory. Resistance

exercise in individuals with and without cardiovascular disease: benefits, rationale,

safety, and prescription: An advisory from the Committee on Exercise,

Rehabilitation, and Prevention, Council on Clinical Cardiology, American Heart

Association; Position paper endorsed by the American College of Sports Medicine.

Circulation 2000; 101:828-833.

15. Mesa JL, Ruiz JR, Ortega FB, et al. Aerobic physical fitness in relation to blood

lipids and fasting glycaemia in adolescents: influence of weight status. Nutr Metab

Cardiovasc Dis 2006; 16:285-293.

16. Ruiz JR, Ortega FB, Rizzo NS, et al. High Cardiorespiratory Fitness is Associated

with Low Metabolic Risk Score in Children; The European Youth Heart Study.

Pediatrics Res In press.

17. Ruiz JR, Sola R, Gonzalez-Gross M, et al. Cardiovascular fitness is negatively

associated with homocysteine levels in female adolescents. Arch Pediatr Adol Med

In press.

18. Gonzalez-Gross M, Castillo MJ, et al. [Feeding and assessment of nutritional status

of spanish adolescents (AVENA study). Evaluation of risks and interventional

proposal. I. Methodology]. Nutr Hosp 2003; 18:15-28.

19. Moreno LA, Joyanes M, Mesana MI, et al. Harmonization of anthropometric

measurements for a multicenter nutrition survey in Spanish adolescents. Nutrition

2003; 19:481-486.

20. Ruiz JR, FB Ortega, LA Moreno, et al. Serum lipid and lipoprotein reference values

of Spanish adolescents; The AVENA study. Soz Praventivmed 2006; 51:99-109.

21. Moreno LA, Mesana MI, Fleta J, et al. Overweight, obesity and body fat

composition in spanish adolescents. The AVENA Study. Ann Nutr Metab 2005;

49:71-76.

22. Moreno LA, Mesana MI, Gonzalez-Gross M, et al. Anthropometric body fat

composition reference values in Spanish adolescents. The AVENA Study. Eur J Clin

Nutr. 2006; 60:191-196.

23. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for

child overweight and obesity worldwide: international survey. BMJ 2000; 320:1240-

1243.

24. Tanner JM, Whitehouse RH. Clinical longitudinal standards for height, weight,

height velocity, weight velocity, and stages of puberty. Arch Dis Child 1976;

51:170-179.

25. Warnberg J, Nova E, Moreno LA, et al. Inflammatory proteins are related to total

and abdominal adiposity in a healthy adolescent population: the AVENA Study. Am

J Clin Nutr 2006; 84: 505-512.

26. Léger LA, Mercier D, Gadoury C, Lambert J. The multistage 20 metre shuttle run

test for aerobic fitness. J Sports Sci 1988; 6:93-101.

27. Committee of Experts on Sports Research EUROFIT. Handbook for the EUROFIT

Tests of Physical Fitness. Strasburg, GE: Council of Europe, 1993.

28. Ortega FB, Ruiz JR, Castillo MJ, et al. [Low level of physical fitness in Spanish

adolescents. Relevance for future cardiovascular health (AVENA study)] Rev Esp

Cardiol 2005; 58: 898-909.

29. Holm S. A simple sequentially rejective multiple test procedure. Scand J Statist

1979; 6:65-70.

30. Jurca R, Lamonte MJ, Church TS, et al. Associations of muscle strength and fitness

with metabolic syndrome in men. Med Sci Sports Exerc 2004; 36:1301-1307.

31. Jurca R, Lamonte MJ, Barlow CE, Kampert JB, Church TS, Blair SN. Association

of muscular strength with incidence of metabolic syndrome in men. Med Sci Sports

Exerc 2005; 37:1849-1855.

32. LaMonte MJ, Eisenman PA, Adams TD, Shultz BB, Ainsworth BE, Yanowitz FG.

Cardiorespiratory fitness and coronary heart disease risk factors: the LDS Hospital

Fitness Institute cohort. Circulation 2000; 102: 1623-1628.

33. Ridker PM, Stampfer MJ, Rifai N. Novel risk factors for systemic atherosclerosis: a

comparison of C-reactive protein, fibrinogen, homocysteine, lipoprotein(a), and

standard cholesterol screening as predictors of peripheral arterial disease. JAMA

2001; 285:2481-2485.

34. Dernellis J, Panaretou M. Effects of C-reactive protein and the third and fourth

components of complement (C3 and C4) on incidence of atrial fibrillation. Am J

Cardiol 2006; 97:245-248.

35. Wellen KE, Hotamisligil GS. Inflammation, stress, and diabetes. J Clin Invest 2005;

115:1111-1119.

36. Cesari M, Penninx BW, Pahor M, et al. Inflammatory markers and physical

performance in older persons: the InCHIANTI study. J Gerontol A Biol Sci Med Sci

2004; 59:242-248.

37. Penninx BW, Kritchevsky SB, Newman AB, et al. Inflammatory markers and

incident mobility limitation in the elderly. J Am Geriatr Soc 2004; 52:1105-1113.

38. Ferrucci L, Penninx BW, Volpato S, et al. Change in muscle strength explains

accelerated decline of physical function in older women with high interleukin-6

serum levels. J Am Geriatr Soc 2002; 50:1947-1954.

39. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to

inflammation. N Engl J Med 1999; 340:448-454.

40. Lennmarken C, Sandstedt S, Schenck HV, Larsson J. The effect of starvation on

skeletal muscle function in man. Clin Nutr 1986; 5: 99-103.

41. Spiekerman AM: Proteins used in nutritional assessment. Clin Lab Med 1993;

13:353-369.

42. Wells JC, Fewtrell MS, Williams JE, Haroun D, Lawson MS, Cole TJ. Body

composition in normal weight, overweight and obese children: matched case-control

analyses of total and regional tissue masses, and body composition trends in relation

to relative weight. Int J Obes 2006; 30:1506-1513.

43. Deforche B, Lefevre J, De Bourdeaudhuij I, Hills AP, Duquet W, Bouckaert J.

Physical fitness and physical activity in obese and nonobese Flemish youth. Obes

Res 2003; 11:434-441.

44. Sothern MS, Loftin JM, Udall JN, et al. Safety, feasibility, and efficacy of a

resistance training program in preadolescent obese children. Am J Med Sci 2000;

319:370-375.

45. Winett RA, Carpinelli RN. Potential health-related benefits of resistance training.

Prev Med 2001; 33:503-513.

46. Shaibi GQ, Cruz ML, Ball GD, et al. Effects of resistance training on insulin

sensitivity in overweight Latino adolescent males. Med Sci Sports Exerc 2006;

38:1208-1215.

47. Olson TP, Dengel DR, Leon AS, Schmitz KH. Moderate resistance training and

vascular health in overweight women. Med Sci Sports Exerc 2006; 38:1558-1564.

48. Fenicchia LM, Kanaley JA, Azevedo JL, et al. Influence of resistance exercise

training on glucose control in women with type 2 diabetes. Metabolism 2004;

53:284-289.

49. Ibanez J, Izquierdo M, Arguelles I, et al. Twice-weekly progressive resistance

training decreases abdominal fat and improves insulin sensitivity in older men with

type 2 diabetes. Diabetes Care 2005; 28:662-667.

50. Ruiz JR, Ortega FB, Gutierrez A, Sjöström M, Castillo MJ. Physical fitness

assessment in childhood and adolescence; A European approach based on the

AVENA, EYHS and HELENA studies. J Public Health 2006; 14:269-277.

Table 1. Physical characteristics of the subjects1

1Values are mean ± SD, unless stated.

2P < 0.05, 3P < 0.01, 4P < 0.001 for gender comparisons adjusted for mass

significance (29). 5Analyses were performed on log-transformed data, but non

transformed data are presented in the table. 6Five categories: low, medium-low,

medium, medium-high and high socioeconomic status.

All (n = 416) Boys (n= 232) Girls (n= 186)

Age (y) 15.4 ± 1.4 15.4 ± 1.4 15.4 ± 1.4

Weight (kg) 61.3 ± 12.8 64.9 ± 13.4 56.8 ± 10.54

Height (cm) 166.6 ± 8.7 171.0 ± 8.1 161.4 ± 6.24

Body mass index (kg/m2) 22.0 ± 3.7 22.1 ± 3.9 21.8 ± 3.6

Overweight (including obesity) (%) 26 31 20†

Pubertal status I/II/III/IV/V (%) 0/3/12/47/38 0/4/15/41/40 0/1/9/55/352

Sum of six skinfolds (mm) 44.5 ± 5.8 43.0 ± 6.1 44.1 ± 5.3

C-reactive protein (mg/L)5 1.44 ± 3.14 1.56 ± 2.61 1.28 ± 3.69

C3 (g/L)5 1.35 ± 0.24 1.36 ± 0.24 1.33 ± 0.22

C4 (g/L)5 0.27 ± 0.10 0.27 ± 0.09 0.27 ± 0.10

Ceruloplasmin (g/L)5 0.21 ± 0.05 0.20 ± 0.04 0.22 ± 0.053

Transthyretin (mg/dL)5 23.76 ± 6.56 24.30 ± 6.90 23.02 ± 6.02

Cardiorespiratory fitness (stages) 5.8 ± 2.8 7.1 ± 2.6 4.1 ± 1.94

Handgrip strength (kg) 31.8 ± 8.0 35.5 ± 7.6 25.4 ± 4.04

Standing broad jump, cm 173.9 ± 32.5 191.0 ± 29.1 152.7 ± 22.24

Socioeconomic status (%)6 5/26/40/23/6 4/27/43/21/5 6/25/37/25/7

Ta

ble

2. A

ssoc

iati

ons

betw

een

infl

amm

ator

y pr

otei

ns a

nd m

uscl

e st

reng

th.

Dat

a w

ere

anal

yzed

by

one-

way

ana

lysi

s of

cov

aria

nce

afte

r ad

just

ing

for

sex,

age

, pub

erta

l st

atus

, wei

ght,

hei

ght,

soc

ioec

onom

ic s

tatu

s, a

nd

card

iore

spir

ator

y fi

tnes

s.

1 All

ana

lyse

s w

ere

perf

orm

ed o

n lo

g-tr

ansf

orm

ed d

ata.

M

uscl

e st

reng

th

Dep

ende

nt v

aria

ble1

Est

imat

ed m

ean

diff

eren

ces

1st t

erti

le -

3rd

ter

tile

P

Est

imat

ed m

ean

diff

eren

ces

2nd t

erti

le -

3rd

ter

tile

PP

for t

rend

C-r

eact

ive

prot

ein

0.35

1 0.

015

0.29

2 0.

032

0.03

0

C3

0.03

8 0.

103

-0.0

08

0.71

30.

103

C4

0.08

5 0.

117

0.11

1 0.

029

0.07

1

Cer

ulop

lasm

in

0.07

0 0.

029

0.00

9 0.

759

0.05

2

Tan

sthy

reti

n -0

.173

0.

292

-0.0

87

0.56

70.

253

Table 3. Associations between inflammatory proteins and body mass index.

Body mass index

Dependent variable3Estimated mean

differences1

N-OVER - OVER P

Estimated mean differences2

N-OVER - OVER P

C-reactive protein -0.432 0.001 -0.236 0.225

C3 -0.095 <0.001 0.025 0.416

C4 -0.221 <0.001 -0.146 0.043

Ceruloplasmin -0.049 0.099 0.027 0.519

Transthyretin -0.067 0.653 -0.196 0.375

1Data were analyzed by one-way analysis of covariance after adjusting for sex, age,

pubertal status, socioeconomic status, and cardiorespiratory fitness, and 2with an

additional adjustment made for skinfold thickness.

N-OVER indicates non-overweight.

3All analyses were performed on log-transformed data.

FIGURE 1. Association between inflammatory proteins and muscle strength tertiles

(first, second and third tertile equals to low, middle and high) in overweight (grey

columns) and non-overweight (white columns) adolescents. Columns are estimated

means. Data were analysed by one-way analysis of covariance separately in

overweight and non-overweight adolescents after adjusting for sex, age, pubertal

status, socioeconomic status, and cardiorespiratory fitness. Absence of P values

indicates no statistically significant association. Ln indicates logarithmic

transformation.

FIGURE 1.

Inflammatory proteins and muscle strength tertiles by body mass index categories

0.0

0.1

0.2

0.3

0.4

0.5

Low Middle High

Ln

C3

0.0

0.1

0.2

0.3

0.4

0.5

Low Middle High

Ln

C4

0.0

0.1

0.2

0.3

0.4

0.5

0.6

Low Middle High

Ln

Ce

rulo

pla

smin P for trend = 0.058

0.1

0.3

0.5

0.7

0.9

1.1

1.3

1.5

Ln

C-r

ea

ctiv

e p

rote

in

Low Middle High

P for trend = 0.039

4.0

4.2

4.4

4.6

4.8

5.0

5.2

5.4

Low Middle High

P for trend = 0.004

Ln

tra

nsth

yre

tin

USE OF ARTIFICIAL NEURAL NETWORK-BASED EQUATION

FOR ESTIMATING VO2MAX IN ADOLESCENTS

Jonatan R. Ruiz1,2, Jorge Ramírez-Lechuga3, Francisco B.

Ortega1,2, Jose M. Benítez4, Antonio Araúzo-Azofra4, Cristóbal

Sanchez3, Michael Sjöström2, Manuel J. Castillo1, Ángel

Gutiérrez1, Mikel Zabala3, on behalf of the HELENA Study Group

Submitted


Granada, Spain. 2 Unit for Preventive Nutrition, Department of Biosciences and

Nutrition at NOVUM, Karolinska Institutet, Huddinge, Sweden. 3 Department of

Physical Education, School of Sport Sciences, University of Granada,

Granada, Spain. 4 Department of Computer Sciences and Artificial Intelligence,

School of Informatics, University of Granada, Granada, Spain.

VII

SUBMITTED

Use of Artificial Neural Network-Based Equation for Estimating VO2max

in Adolescents

RUIZ, JONATAN R.1,2*; RAMIREZ-LECHUGA, JORGE 3*; ORTEGA,

FRANCISCO B. 1,2; BENITEZ, JOSE M.4; ARAUZO-AZOFRA, ANTONIO 4;

SANCHEZ, CRISTOBAL3, SJÖSTRÖM, MICHAEL2; CASTILLO, MANUEL J. 1;

GUTIERREZ, ANGEL1; ZABALA, MIKEL3 ON BEHALF OF THE HELENA

STUDY GROUP

1 Department of Physiology, School of Medicine, University of Granada, Granada,

Spain. 2 Unit for Preventive Nutrition, Department of Biosciences and Nutrition at

NOVUM, Karolinska Institutet, Huddinge, Sweden. 3 Department of Physical Education, School of Sport Sciences, University of

Granada, Granada, Spain. 4 Department of Computer Sciences and Artificial Intelligence, School of

Informatics, University of Granada, Granada, Spain.

* Contributed equally.

Running title: Neural network-equation for estimating VO2max

Address for correspondence: Jonatan R. Ruiz. Department of Physiology, School of

Medicine, University of Granada, 18071, Granada, Spain. Tel: +34 958 243 540,

Fax: + 34 958 249 015, E-mail: [email protected]

ABSTRACT

Purpose: To develop an artificial neural network (ANN)-equation to estimate

maximal oxygen uptake (VO2max) from 20m shuttle run test (20mSRT) performance

(stage), sex, age, weight and height in young persons.

Methods: The 20mSRT was performed by 193 (122 boys and 71 girls) adolescents

aged 13-19 years. All the adolescents wore a portable gas analyzer to measure VO2

and heart rate during the test. The equation was developed and cross-validated

following the ANN mathematical model. The neural net performance was assessed

through several error measures. Agreement between the measured VO2max and

estimated VO2max from Léger’s and ANN equations were analysed following the

Bland and Altman method.

Results: The percentage error was 17.13 and 7.38 for Léger and ANN-equation,

respectively, and the standard error of the estimate obtained with Léger’s equation

was 4.27 ml/kg/min, while for the ANN equation was 2.84 ml/kg/min. A Bland-

Altman plot for the measured VO2max and Léger-VO2max showed a mean difference of

4.9 ml/kg/min (P < 0.001), while the Bland-Altman plot for the measured VO2max

and ANN-VO2max showed a mean difference of 0.5 ml/kg/min (P = 0.654).

Conclusions: In this study, an ANN-based equation to estimate VO2max from

20mSRT performance (stage), sex, age, weight and height in adolescents was

developed and cross-validated. The newly developed equation was shown to be more

accurate than Léger’s equation in the sample of adolescents studied. The proposed

model has been coded in a user friendly spread sheet.

Key words: Cardiorespiratory fitness; maximal oxygen uptake; aerobic capacity test;

exercise field test.

INTRODUCTION

The maximal rate of oxygen uptake (VO2max) is considered as a gold standard for

measurement of cardiorespiratory fitness. Cardiorespiratory fitness is a direct

marker of physiological status and reflects the overall capacity of the cardiovascular

and respiratory systems and the ability to carry out prolonged exercise (35). In

addition, recent reports suggest that cardiorespiratory fitness is also an important

health marker in young persons. High cardiorespiratory fitness during childhood and

adolescence has been associated with a favourable plasma lipid profile in both

overweight and non-overweight adolescents (18), with total body fat (29), features of

the metabolic syndrome (4, 28), novel cardiovascular disease risk factors (30), and

with arterial compliance (26) in young people. These findings support the concept

that cardiorespiratory fitness may exert a protective effect on the cardiovascular

system from an early age.

Cardiorespiratory fitness is one of the main health-related physical fitness

components used in schools, sports centres and health institutions. One of the most

widely used field tests for estimating cardiorespiratory fitness among adolescents is

the 20m shuttle run test (20mSRT) also called the “Course Navette” test (12, 37).

The 20mSRT or a slight modification of it, has been included in several fitness

batteries, such as the EUROFIT (5), and the American Progressive Aerobic

Cardiovascular Endurance Run and the FITNESSGRAM battery (36) among others.

The 20mSRT is a feasible fitness test, since a large number of subjects can be tested

at the same time, which enhances the motivation of the participants. It can be

conducted indoors or outdoors in a relatively small area, and on different surfaces

(slippery and rubber floors).

Several equations have been developed to estimate VO2max from maximal speed

attained during the 20mSRT (Table 1). Léger et al. (13) developed an equation based

on a sample of 188 boys and girls aged 8-19 years to estimate the VO2max from

maximal speed attained during the 20mSRT, age and the speed and age interaction.

However, Léger’s equation has some limitations. Sex is not included in the model,

yet it is well known that physical performance is highly different in boys and girls of

all ages. Moreover, the estimates of VO2max for low scores were based on

extrapolated data from the study since the original study population did not have data

for these points. The accuracy of the Léger’s (13) prediction model has been

examined by several researchers (1, 6, 14, 15, 24, 33, 34, 38), but no attempts have

been made to develop a more accurate model in a wide age range.

It seems viable to develop a more accurate VO2max equation for the adolescent

period, while taking those variables which have been shown to have an impact on the

level of cardiorespiratory fitness into account. Published equations for VO2max have

the shape of a linear or quasi-linear expression on different input variables (sex, age,

body weight, and stage) (Table 1). Researchers have used these type of models

mainly because of their simplicity, easy of use, and familiarity. A way forward in

obtaining an improved model could be done by exploring the feasibility of some new

methods. Recently, there has been a growing interest in artificial neural networks

(ANN). ANNs have some theoretical advantages over more traditional regression

methods (7). Predictive models based on ANNs have been studied extensively in

many areas of medicine (e.g. breast cancer diagnosis, mortality assessment in

intensive care units, diagnostic scoring, renal function evaluation).

The aim of this study was to develop an ANN-equation to better estimate VO2max

from 20mSRT performance (stage), sex, age, weight and height in adolescents.

METHODS

Subjects

A total of 203 adolescents (127 boys and 76 girls) aged 13-19 years volunteered to

participate in the study after receiving a detailed explanation about the aim and the

clinical implications of the investigation. A comprehensive verbal description of the

nature and purpose of the study was also given to the teachers. Written informed

consent was obtained from parents, and verbal assent was obtained from participants.

The criteria for inxclusion were: smoking, no personal history of cardiovascular or

metabolic disease, free of disease, any muscular or skeletal injuries, medication at

the time of the study and pregnancy. The experimental protocol was approved by the

Review Committee for Research Involving Human Subjects at the University of

Granada, Spain.

A few adolescents (n = 5) discontinued the test because of discomfort or distress. A

small number of technical problems (n = 5) also occurred, which probably yielded

inaccurate VO2max measurements as a result. Therefore, the final sample was

confined to 193 (122 boys and 71 girls) adolescents with reliable measures of

VO2max.

Procedure

All participants performed the 20mSRT as previously described by Léger et al. (12).

Participants were required to run between two lines 20m apart, while keeping the

pace with audio signals emitted from a pre-recorded CD. The initial speed was 8.5

km/h, which was increased by 0.5 km/h per minute (one minute equal one stage).

The CD used was calibrated over one minute of duration. Participants were

instructed to run in a straight line, to pivot on completing a shuttle, and to pace

themselves in accordance with the audio signals. The test was finished when the

participant failed to reach the end lines concurrent with the audio signals on two

consecutive occasions. Otherwise, the test ended when the subject stopped because

of fatigue. All measurements were carried out under standardized conditions on an

indoor rubber floored gymnasium. The participants were encourage to keep running

as long as possible throughout the course of the test.

All participants were familiar with the test, because the 20mSRT is one of the fitness

tests included in the curriculum of Physical Education in Spain. However, one week

prior the test, participants received a comprehensive instruction after which they also

practiced the test. Subjects were instructed to abstain from strenuous exercises 48

hours prior to the test. All the tests were conducted by the same investigators and at

the same time for each subject (between 10:00 to 13:00 hrs).

Physiological measurements

Heart rate was recorded every 5 seconds throughout the 20mSRT using a Polar

telemetry system (Polar 610i). Moreover, participants wore a portable gas analyzer

(K4b2, Cosmed, Rome, Italy), the purpose of which was to measure the VO2 during

the 20mSRT. Respiratory parameters were recorded breath-by-breath, which in turn

were averaged over a 10 second period. VO2max was the main parameter determined

using the open circuit method. Exhaustion was confirmed when: 1) maximal heart

rate was greater than 185 beats per minute, 2) respiratory exchange ratio was greater

than 1.1, and/or 3) a detection of a plateau in the VO2 curve, defined as an increase of

VO2 less than 2 ml/kg/min with a concomitant increase in stage.

The weight of the Cosmed K4b2 equipment is 1.5 kg including the battery and a

specially designed harness. It has been proven to be a valid device when compared

with the Douglas bag method (17). Wearing the portable gas analyzer during the

20mSRT do not significantly alter the subjects’ energy demands, as it has been

reported (6).

Before each test was conducted, the oxygen and carbon dioxide analyzers were

calibrated according to the manufacturer’s instructions. This consisted of performing

a room air calibration and a reference gas calibration using 15.93 % oxygen and 4.92

% carbon dioxide. The flow turbine was then calibrated using a 3-liter syringe

(Hans-Rudolph). Finally, a delay calibration was performed to adjust for the lag time

that occurs between the expiratory flow measurement and the gas analyzers. During

each test, a gel seal was used to help prevent air leaks from the face mask.

The total time (in seconds) and the last half-stage completed (here called “stage”)

were recorded. The measured VO2max was obtained directly from the K4b2 data.

Estimated VO2max was calculated by the Léger’s equation (13) (Table 1). Height and

body weight was measured to 0.1 kg using a standard beam balance, and body height

was measured to the 0.1 cm using a transportable stadiometer, with the participants

clad only in their underwear. These measures were taken prior the test.

Statistical analyses

The mathematical model used to build a new equation to estimate VO2max from

20mSRT performance (stage), sex, age, weight and height in adolescents was an

ANN. An ANN is a mathematical model that emulates some of the observed

properties of biological nervous systems and draw on the analogies of adaptive

biological learning. The ANN modelling procedure is described in detail elsewhere

(9). Briefly, to solve a problem using ANN, a number of steps must be taken:

1. Select the type of neural net for the type of regression problem that is to be map,

i.e. identification of a VO2max estimator. One of the best options for that purpose is to

use a multilayered perceptron.

2. Data preprocessing. The data gathered for this study consists of a set of 193

instances, each instance being composed of six variables. All variables were

originally expressed in their original units, i.e. sex (boys/girls), age (years), weight

(kilograms), height (centimetres), stage (last-half stage completed), and VO2max

(ml/kg/min). The sample data was afterwards normalized to the [0.1, 0.9] interval,

which simplified the learning of the ANN regression model.

3. Network design. The ANN architecture, i.e. the number of input and output

variables is set by the problem. There are plenty of different models of neural

networks to chose from, each one having its specific properties and advantages for

its particular application. One of the most successful and most popular is the feed-

forward multilayered perceptron. In this network, the computing units are arranged

into three layers, which are conveniently ordered. The information flows forward

from the five neurons of the input layer to the three connecting neurons of the hidden

layer and finally, to the single neuron of the output layer using no backward

connection. The first layer (the input layer) corresponds to the independent variables

(sex, age, weight, height and stage), while the third layer (the output layer)

corresponds to the dependent variable score (VO2max). The intermediate layer, which

is a hidden layer, consisting of all possible connections between the input and the

output layer, allows for a combined impact of a multiple set of independent variables

on the output layer. This would be the same as testing all possible interactions in a

regression model, but without adding any extra degrees of freedom. The neurons in

each layer serve the purpose of optimally transforming each quantitative variable in a

curvilinear fashion, similar to adding a spline function for each of the independent

variables of a regression model. The architecture of the network used in this study is

a multilayered perceptron (5-3-1), which is shown in Figure 1.

4. Find learning algorithm parameters. In order to obtain the synaptic weights of the

ANN, we used the well-known backpropagation algorithm (31). The value for the

algorithm parameters are 0.2 for the learning rate, and 0.5 for the momentum term.

The training of the network is stopped when the sum of squared errors (SSE) falls

below 0.00001 or when 1,500 training epochs have been performed.

5. Training of the network. The ANN-model is identified by means of a data-driven

process, where a fraction of the available data set is used for designing the model and

it is referred to as the training set. The remaining set of data is not used in the design

of the model as such but rather for evaluating its validity once it is ready. This

particular data set is called the test set.

6. Validation of the model. In order to validate the feasibility of the ANN-model for

this problem, a cross-validation technique was applied (19). It means that the total

dataset (composed of 193 samples) was randomly split into k parts with the same

number of samples, except one of them (C = c1, …, ck). The process consists of

building k different neural networks. For the model i, with i = 1,…,k the part ci is

used as the test set, and the remainder (all but ci) are used as the training set. In our

experiments, the value we have used for k is the total number of samples in the data

set (n = 193). Thus each of the nets are built with different training sets, and

evaluated on different and independent test sets. The overall evaluation of the

methodology is measured as the average of the performance on the test sets.

The neural net performance is assessed through an error measure. Suppose that N

cases are available to evaluate the model, where y is the actual output (the measured

VO2max) and ŷ is the output computed by the net (estimated VO2max from the ANN-

equation). Then, a common measure is the SSE defined as:

∑=

−=N

iii yySSE

1

2)ˆ(

An easier way of understanding the expression for the error is to use the percentage

error, which can be computed as follows: First, the SSE is averaged over the number

of cases, rendering the mean sum of squared errors (MSE):

∑=

−=N

iii yy

NMSE

1

2)ˆ(1

MSE is then converted into domain units by taking the root square and yielding the

root mean sum of squared errors (RMSE):

MSERMSE =

The percentage error should intuitively serve as a good indicator of the performance

of a given model:

100thdomain wid

% ×=RMSEError

The standard error of estimate (SEE), is another way to illustrate the performance of

the ANN-model, which also serves for comparative purpose:

)1( 2yyy RSDSEE −=

where SD is the standard deviation of the estimated VO2max from the ANN-model,

and R2 is the correlation between the measured the measured VO2max and the

estimated VO2max from the ANN-model.

The SSE difference between the Léger’s equation and the ANN-model was

examined by paired t test. A second ANN-model was built with the same procedure

and variables as the previous one, but instead of the last half-stage completed, the

last stage completed was used.

Sex differences were analysed by one-way analysis of variance (ANOVA), and

adjusted for mass significance as described by Holm (10). Bivariate correlation

analysis was done in order to examine the relationship between the measured VO2max

and the input variables (age, weight, height and stage) in boys and girls. The

relationship between the measured VO2max and a similar estimated VO2max from

Léger’s equation and the ANN-model was also examined. The overall differences

between the measured VO2max and the similar estimated value from Léger’s equation

and ANN-model was calculated by means of paired t test. The agreement between

the measured VO2max and the similar value as estimated from Léger’s and the ANN

equation was assessed according to the method by Bland and Altman (2, 3).

RESULTS

Physical characteristics and the 20mSRT performance of the participants are

presented in Table 2. Boys and girls were similar in age, but boys were significantly

taller and heavier than girls. Moreover, boys had significantly higher values in all the

20mSRT performance-related variables. A bivariate correlation analysis between the

measured VO2max, age, weight, height and stage in boys and girls is presented in

Table 3. VO2max was significantly associated with age, weight and stage in both

sexes. A borderline significant association was found between VO2max and height in

both boys and girls. Figure 2 shows the relationship between the measured VO2max

and the estimated VO2max from the Léger’s equation, and Figure 3 shows the

relationship between the measured VO2max and the estimated VO2max from the ANN-

equation. Estimated VO2max from both the Léger’s and the ANN-equation were

significantly correlated with the measured VO2max (r = 0.90 and 0.96, respectively,

both P < 0.001).

The evaluation of the error of the VO2max measurements obtained from Léger’s and

the ANN-equation is presented in Table 4. The SSE was significantly higher in

Léger’s equation than in the ANN-equation (P < 0.001), and the percentage error

was 17.13 for the former and 7.38 for the latter. The SEE obtained with the Léger’s

equation was 4.27 ml/kg/min, while for the ANN equation was 2.84 ml/kg/min. The

SSE obtained from the ANN-model built with the last stage completed was

significantly higher than the SSE obtain from the ANN-model built with the last

half-stage completed (1699.48 vs 1600.91 vs, respectively, P = 0.002).

A Bland-Altman plot for the measured VO2max and VO2max estimated from Léger

equation showed a mean difference of 4.9 ml/kg/min (Figure 4). The 95% limits of

agreement ranged from -4.3 to 14.1 ml/kg/min. There was a statistically significant

difference between measured VO2max and Léger equation (47.7 vs 43.0 ml/kg/min, P

< 0.001). A Bland-Altman plot for the measured VO2max and the ANN-equation

showed a mean difference of 0.5 ml/kg/min (Figure 5). The 95% limits of agreement

ranged from -5.1 to 6.1 ml/kg/min. There was no statistical significance difference

between measured VO2max and ANN-equation (47.7 vs 47.2 ml/kg/min, P = 0.654).

DISCUSSION

In this study, an ANN-based equation to estimate VO2max from 20mSRT

performance (stage), sex, age, weight and height in a sample of 193 adolescents aged

13-19 years was developed and cross-validated. The equation is based on: 1) direct

VO2 data collected while the adolescents performed the 20mSRT; 2) the use of a

numerical procedure to build the ANN-equation; 3) a fairly large amount of

adolescents participating in the test; 4) using variables, which have been previously

shown to have an influence on the VO2max for the particular age group being tested.

All variables included in the equation are measured in field studies and no specific

equipment is required to collect the data. All the technical and environmental

variables that may have an influence on the results were carefully controlled in order

to obtain highly reliable VO2 measures; and 5) the use of a precise method for

assessing agreement between two methods. The most frequently used summary

statistics to assess overall agreement between the measurements of different methods

were correlation coefficient. However, correlation is a measure of the strength of

association between two variables but not necessarily a measure of agreement (27).

The ANN-based equation proved to be more accurate for a prediction of the VO2max

value than Léger’s equation for the particular sample of adolescents studied here.

Léger’s equation had an error of 17.13%, while the ANN-equation had an error of

7.38%. The SEE calculated from Léger’s equation was almost twice as high as that

obtained with the ANN-equation (4.27 vs 2.84 ml/kg/min, respectively). Moreover,

Léger’s equation significantly underestimated VO2max by 4.9 ml/kg/min when

compared with the measured VO2max (P < 0.001), while the ANN-equation slightly

underestimated VO2max by 0.5 ml/kg/min (P = 0.654). These results of this study are

in alignment with previous research, which has shown a systematic underestimation

of the VO2max value calculated from Léger’s equation (32, 33).

Differences between the results obtained from the ANN-equation and those obtained

from Léger’s equation may be partly explained by the test protocols and the gas

analysis procedures used for the tests. Léger et al. recorded VO2max by using the

backward extrapolation technique (13). This technique has been extensively

validated, but it can only be considered as an estimate of the actual VO2max. The

present method seems to be a more sensitive method, since data were averaged every

10 seconds, which allowed for the detection of a plateau in the VO2 over the final

workloads.

The ANN-equation has other advantages over Léger’s equation, and also on more

recently published regression equations (Table 1). The reason why sex, age, weight,

height and stage were used as predictive input variables for estimating VO2max in the

ANN-equation is reviewed below.

Sex. As it could be expected, there was a significant difference between boys and

girls in the measured VO2max value. This result is also consistent with normative data

showing lower levels of VO2max for girls than for boys (23). However, Léger’s

equation does not account for sex. Factors explaining the lower VO2max values for

girls may be partially explained by the fact that girls usually have a lower

development of muscular mass and higher fraction of body fat (20). Moreover, it has

been suggested that women may be more prone to pulmonary limitations during

heavy exercise (and perhaps submaximal intensities) than men, which is supposedly

due to the influence of the reproductive hormones (estrogen and progesterone) in

combination with a reduced pulmonary capacity (8). A greater ventilatory work

associated with an increased expiratory flow limitation during the exercise and gas

exchange impairments seems to be of primary importance. The influence of sex on

VO2max has also been taken into account in two recently published equations (15,

33). Sticklan et al. (33) developed two sex-specific equations with similar slopes for

both men and women aged 18-38 years. They found a slightly lower Y-intercept

value for women, which is in agreement with our findings. Mahar et al. (15)

developed an equation based on a sample consisting of 61 boys and 74 girls in the

age group between 12-14 years in which sex, number of laps completed, and body

weight were included as independent variables (Table 1).

Age. Léger et al. (12) included age as one of the independent variables in their

model, which was not the case in other published equations (6, 15, 33). The age

range of the adolescents involved in the present study was similar to the study made

by Léger et al. (12). However, the youngest adolescent in our study was 13 years old,

while the youngest person in Léger’s study was 8 years old. Findings from cross-

sectional and longitudinal studies have shown that age is associated with VO2max in

both adolescents and adults (23).

Adolescence represents a period of life where changes such as growth and

physiological development occur. Therefore, age might be an important factor to

check for in order to understand the contribution of those age dependent factors. It

has been suggested that rather than using the chronological age as a measure for this

variable, sexual maturation (i.e. biological age) may be a more accurate marker of

the physiological status of the person in this period of her/his life (25).

However, findings from cross-sectional studies examining the influence of sexual

maturation on VO2max have shown that sexual maturation may account for only a

small proportion of the variance in the measured VO2max value (21), and that weight

and height are primarily responsible for variation in VO2max throughout maturation

(11). One of the main reasons why sexual maturation was not included in the

equation was due to the suspected inaccuracy in self-reporting tanner stage in some

circumstances, and the need for a paediatrician or trained physician to make an

objective measurement, which in most setting is not feasible.

Body size: weight and height. The increases in VO2max are influenced by changes in

body weight and height. Controlling the effect of a body size changes in growing

adolescents is critical in order to understand the relative contributions of other

factors influencing changes in the value of VO2max, such as sex, maturation, habitual

physical activity, and functional cardiorespiratory improvements. The conventional

(ratio) approach for controlling or, "normalizing" VO2max for body size has been to

divide the VO2max value by kilogram of body weight. However, in walking/running

activities, height also has been shown to have an impact on the performance, and

specially in those activities incorporating shuttle running such as the 20mSRT (6,

16). Body weight has usually been used as a measure of body size, but it has also

been suggested that height could be used when scaling body size to account for

possible disproportionate changes in muscle mass with increasing body size (40).

This study shows that both body weight and height are significantly correlated with

the 20mSRT performance (Table 3). Body weight was negatively correlated with

VO2max in both sex (r = -0.517, P < 0.001; r = -0.241, P < 0.043 for boys and girls,

respectively), while the correlation between height and VO2max was less evident for

both sex (r = -0.170, P = 0.079; r = 0.219, P = 0.066 for boys and girls,

respectively). It is worth noting that height is negatively correlated with VO2max for

boys, while the opposite is true for girls. Girls had significantly lower height than

boys (161.0 vs 172.5 cm, respectively, P < 0.001), which may indicate that height

has a positive contribution on the 20mSRT performance up to a certain level after

which it has a negative impact. It is tenable that various biomechanical complexities

of shuttle running may account for this. Other approaches have recommended the use

of allometric scaling exponents (39) or accounting for fat free mass (22) in order to

allow for a more appropriate study on the impact of body size differences on VO2max.

However, the allometric scaling exponents has not been universally reported (40),

and the use of fat free mass needs either expensive instrumentation or trained

evaluators (when derived from anthropometric measurements) which is not often a

feasible choice, especially in schools settings.

To our knowledge there is only one model that includes body size measurements in

the model (15). The equation developed by Mahar et al. (15) includes both weight

and height as single variables and as a ratio [body weight in kilogram divided by

height in meters squared (BMI)] (Table 1). They also developed another equation

where only weight is used as a predictive variable in the model. A SEE of 6.38 and

6.35 ml/kg/min was reported for the first and second equation, respectively. These

results are slightly higher than the SEE obtained in the present study by means of

both the Léger’s and the ANN-equation (4.27 and 2.84 ml/kg/min, respectively).

Some aspects of the used methods could probably explain the observed differences

in the SEE values. Mahar et al. (15) used a multiple regression model to predict the

measured VO2max from the number of laps completed on the 20mSRT. The following

variables were included: sex and body mass or BMI. The dependent variable in the

regression model was measured VO2max, which was collected while running until

exhaustion on the treadmill. Energy demands during shuttle running have been

reported to be higher when compared with treadmill running (6), which can be

attributed to the mode of exercise, technique, and musculature employed in the two

conditions. This may confound the reliability of the equations built with VO2max

values collected from treadmill-based protocols (6, 15, 33).

Stage. In the ANN-equation, the maximal 20mSRT performance attained is

calculated from the last half-stage completed, so it allows credit for 30 second when

participants fall short of completing a full stage. This increased precision should help

in detecting changes in fitness in interventional studies, follow up studies, in athletes

before and after a period of training, etc. The Léger’s equation used maximum speed

calculated from the last stage completed, and therefore, subjects falling just short of

completing a full one-minute stage would be ascribed to the previously completed

stage. Consequently, the ANN-equation may allow for a greater sensitivity in the

estimation of VO2max when compared with Léger’s equation. Stickland et al. (33)

also used the last half-stage completed as the measure of the 20mSRT performance

to build a prediction model for adults, and it allowed for a higher degree of accuracy

when compared with Léger’s equation.

Constraints. It is important to acknowledge that the 20mSRT is a test requiring

maximal effort. Special attention has to be paid during the course of the test as such,

since today there are at least three major variants of the test available. Special

attention should also be on the cassette or CDs to be used. Methodological variations

in these cassettes (e.g. calling the stage number at the start versus the finish of each

stage; using only full minutes versus both full minutes and half minutes to indicate

completed stages) will mean that identical performances are reported in different

ways.

The main limitation of the ANN is its complexity and its “black box” nature. The

complexity of the ANN-equation may become rather inconvenient when applied in

the field. However, even when using Léger’s equation in a relatively big sample of

subjects, a programmable device (spreadsheet) is required. Similarly, the estimation

of VO2max using the proposed ANN-equation can be done by means of a spreadsheet.

Some of the advantages of using an ANN-model will need some special attention: 1)

Its capability of producing a nonlinear input-output mapping. A neural network

computes a function, which maps its inputs variables with its output. A nonlinear

relationship could exist between the input and the output variable. However, ANNs

are especially suitable for modelling highly nonlinear maps; 2) its learning ability

(adaptivity). A neural network can be trained to perform a specific task, for example,

reproducing an unknown input-output mapping. There is always a neural network

which will match your input variables as closely as possible with your output for a

given set of data. In other words, you can approximate a given input-output map

with a network as precise as you need; and 3) the ability to generalize. An ANN-

model can be set up to be trained to produce a correct output for a given set of input

data. The applications from the present investigation would be further increased by

performing validation studies in specific populations and in different countries.

In conclusion, in this study an ANN-equation to estimate VO2max from 20mSRT

performance (stage), sex, age, weight and height in adolescents aged 13-19 years was

developed and cross-validated. The newly developed equation was shown to be more

accurate than Léger’s equation in the sample of adolescents studied. All variables

included in the equation are usually measured in field studies, no specific equipment

is required to collect the data, and is not time-consuming. The proposed model has

been coded in a user friendly spread sheet.

ACKNOWLEDGEMENTS

The present paper is published on behalf of the HELENA (Healthy Lifestyle in

Europe by Nutrition in Adolescence) Study group

(http://www.helenastudy.com/list.php). The HELENA Study takes place with the

financial support of the European Community Sixth RTD Framework Programme

(Contract FOOD-CT-2005-007034). The content of this article reflects only the

authors’ views and the European Community is not liable for any use that may be

made of the information contained therein. This study was also partially supported

by Ministerio de Ciencia y Teconología (TIC2003-04650, TIN2004-07236) and

FEDER funds (ERGOLAB-project CIT300100-2005-23). JRR and FBO were

supported by a grant from Ministerio de Educación y Ciencia de España (AP2003-

2128, AP2004-2745).

No benefits in any form have been received or will be received from a commercial

party related directly or indirectly to the subject of this article. None of the authors

had any conflict of interest. The results of the present study do not constitute

endorsement of the product by the authors or ACSM.

REFERENCES 1. Anderson, G. S. The 1600-m run and multistage 20-m shuttle run as predictive tests

of aerobic capacity in children Ped Exerc Sci. 4:312-318 1992.

2. Bland, J. M. and D. G. Altman. Comparing methods of measurement: why plotting

difference against standard method is misleading. Lancet. 346:1085-1087 1995.

3. Bland, J. M. and D. G. Altman. Statistical methods for assessing agreement between

two methods of clinical measurement. Lancet. 8:307-310, 1986.

4. Brage, S., N. Wedderkopp, U. Ekelund, P. W. Franks, N. J. Wareham, L. B.

Andersen, and K. Froberg. Features of the metabolic syndrome are associated with

objectively measured physical activity and fitness in Danish children: the European

Youth Heart Study (EYHS). Diabetes Care. 27:2141-2148, 2004.

5. Council of Europe committee for the development of sport. Eurofit. Handbook for

the EUROFIT tests of physical fitness. Rome (Italy): Edigraf editoriale grafica,

1988, 19-37.

6. Flouris, A. D., G. S. Metsios, and Y. Koutedakis. Enhancing the efficacy of the 20 m

multistage shuttle run test. Br J Sports Med. 39:166-170, 2005.

7. Goodman, C. S. and R. Ahn. Methodological approaches of health technology

assessment. Int J Med Inform. 56:97-105, 1999.

8. Harms, C. A. Does gender affect pulmonary function and exercise capacity? Respir

Physiol Neurobiol. 151:124-131, 2006.

9. Haykin, S. Neural Networks, a Comprehensive Foundation: Prentice-Hall, 1999, 20-

60.

10. Holm, S. A simple sequentially rejective multiple test procedure Scand J Statist.

6:65-70, 1979.

11. Jones, M. A., P. J. Hitchen, and G. Stratton. The importance of considering

biological maturity when assessing physical fitness measures in girls and boys aged

10 to 16 years. Ann Hum Biol. 27:57-65, 2000.

12. Léger, L. A., A. Lambert, A. Goulet, C. Rowan, and Y. Dinelle. [Aerobic capacity

of children aged 6 to 17 years- test Navette of 20m with stages of 1 minute]. Can J

Appl Sport Sci. 9:64-69, 1984.

13. Leger, L. A., D. Mercier, C. Gadoury, and J. Lambert. The multistage 20 metre

shuttle run test for aerobic fitness. J Sports Sci. 6:93-101, 1988.

14. Liu, N. Y., S. A. Plowman, and M. A. Looney. The reliability and validity of the 20-

meter shuttle test in American students 12 to 15 years old. Res Q Exerc Sport.

63:360-365, 1992.

15. Mahar, M. T., G. J. Welk, D. A. Rowe, D. J. Crotts, and K. L. McIver. Development

and Validation of a Regression Model to Estimate VO2peak From PACER 20-m

Shuttle Run Performance. J Phys Act Health:S34-S46, 2006.

16. McCann, D. J. and W. C. Adams. The size-independent oxygen cost of running.

Med Sci Sports Exerc. 35:1049-1056, 2003.

17. McLaughlin, J. E., G. A. King, E. T. Howley, D. R. Bassett, Jr., and B. E.

Ainsworth. Validation of the COSMED K4 b2 portable metabolic system. Int J

Sports Med. 22:280-284, 2001.

18. Mesa, J. L., J. R. Ruiz, F. B. Ortega, J. Warnberg, D. Gonzalez-Lamuno, L. A.

Moreno, A. Gutierrez, and M. J. Castillo. Aerobic physical fitness in relation to

blood lipids and fasting glycaemia in adolescents: influence of weight status. Nutr

Metab Cardiovasc Dis. 16:285-293, 2006.

19. Mitchell, T. Machine Learning: WCB/McGraw-Hill, 1997, 17-45.

20. Moreno, L. A., M. I. Mesana, M. Gonzalez-Gross, C. M. Gil, J. Fleta, J. Warnberg,

J. R. Ruiz, A. Sarria, A. Marcos, and M. Bueno. Anthropometric body fat

composition reference values in Spanish adolescents. The AVENA Study. Eur J

Clin Nutr. 60:191-196, 2006.

21. Mota, J., S. Guerra, C. Leandro, A. Pinto, J. C. Ribeiro, and J. A. Duarte.

Association of maturation, sex, and body fat in cardiorespiratory fitness. Am J Hum

Biol. 14:707-712, 2002.

22. Neder, J. A., M. C. Lerario, M. L. Castro, A. Sachs, and L. E. Nery. Peak VO2

correction for fat-free mass estimated by anthropometry and DEXA. Med Sci Sports

Exerc. 33:1968-1975, 2001.

23. Ortega, F. B., J. R. Ruiz, M. J. Castillo, L. A. Moreno, M. Gonzalez-Gross, J.

Warnberg, and A. Gutierrez. [Low level of physical fitness in Spanish adolescents.

Relevance for future cardiovascular health (AVENA study)]. Rev Esp Cardiol.

58:898-909, 2005.

24. Pitetti, K. H., B. Fernhall, and S. Figoni. Comparing two regression formulas that

predict VO2 peak using the 20-m shuttle run for children and addolescents. Ped

Exerc Sci:125-134, 2002.

25. Preece, M. A. Evaluation of growth and development. In: Avner, ED, Barratt TM,

Holliday MA, eds. Pediatric Nephrology. 3rd ed. Philadel-phia: Williams & Wilkins,

1994, 378-396.

26. Reed, K. E., D. E. Warburton, R. Z. Lewanczuk, M. J. Haykowsky, J. M. Scott, C.

L. Whitney, J. M. McGavock, and H. A. McKay. Arterial compliance in young

children: the role of aerobic fitness. Eur J Cardiovasc Prev Rehabil. 12:492-497,

2005.

27. Rothwell, P. M. Analysis of agreement between measurements of continuous

variables: general principles and lessons from studies of imaging of carotid stenosis.

J Neurol. 247:825-834, 2000.

28. Ruiz, J. R., F. B. Ortega, N. S. Rizzo, L. Villa, A. Hurtig-Wennlöf, L. Oja, and M.

Sjöström. High Cardiorespiratory Fitness is Associated with Low Metabolic Risk

Score in Children; The European Youth Heart Study. Pediatrics Res, In press.

29. Ruiz, J. R., N. S. Rizzo, A. Hurtig-Wennlof, F. B. Ortega, J. Warnberg, and M.

Sjostrom. Relations of total physical activity and intensity to fitness and fatness in

children: the European Youth Heart Study. Am J Clin Nutr. 84:299-303, 2006.

30. Ruiz, J. R., R. Sola, M. Gonzalez-Gross, F. B. Ortega, G. Vicente-Rodriguez, M.

Garcia-Fuentes, A. Gutierrez, M. Sjostrom, K. Pietrzik, and M. J. Castillo.

Cardiovascular fitness is negatively associated with homocysteine levels in female

adolescents. Arch Pediatr Adol Med, In press.

31. Rumelhart, D. E., G. E. Hinton, and R. J. Williams. Learning Representations of

Back-propagation Errors. Nature. 323:533-536, 1985.

32. St Clair Gibson, A., S. Broomhead, M. I. Lambert, and J. A. Hawley. Prediction of

maximal oxygen uptake from a 20-m shuttle run as measured directly in runners and

squash players. J Sports Sci. 16:331-335, 1998.

33. Stickland, M. K., S. R. Petersen, and M. Bouffard. Prediction of maximal aerobic

power from the 20-m multi-stage shuttle run test. Can J Appl Physiol. 28:272-282,

2003.

34. Suminski, R. R., N. D. Ryan, C. S. Poston, and A. S. Jackson. Measuring aerobic

fitness of Hispanic youth 10 to 12 years of age. Int J Sports Med. 25:61-67, 2004.

35. Taylor, H. L., E. Buskirk, and A. Henschel. Maximal oxygen intake as an objective

measure of cardio-respiratory performance. J Appl Physiol. 8:73-80, 1955.

36. The Cooper Institute for Aerobics Research. FITNESSGRAM test administration

manual. Champaign, IL: Human Kinetics, 1999, 38-39.

37. Tomkinson, G. R., L. A. Leger, T. S. Olds, and G. Cazorla. Secular trends in the

performance of children and adolescents (1980-2000): an analysis of 55 studies of

the 20m shuttle run test in 11 countries. Sports Med. 33:285-300, 2003.

38. Van Mechelen, W., H. Hlobil, and H. C. G. Kemper. Validation of two running tests

as estimates of maximal aerobic power in children. Eur J Appl Physiol Occup

Physiol. 55:503-506, 1986.

39. Vanderburgh, P. M., M. T. Mahar, and C. H. Chou. Allometric scaling of grip

strength by body mass in college-age men and women. Res Q Exerc Sport. 66:80-84,

1995.

40. Welsman, J. R., N. Armstrong, A. M. Nevill, E. M. Winter, and B. J. Kirby. Scaling

peak VO2 for differences in body size. Med Sci Sports Exerc. 28:259-265, 1996.

TA

BL

E 1

. Equ

atio

ns to

est

imat

e V

O2m

ax.

Stud

y Sa

mpl

e A

ge (y

) In

put v

aria

bles

E

quat

ion

to e

stim

ate

VO

2max

(ml/k

g/m

in)

Lége

r et a

l. (2

1)

188

boys

&

Girl

s 8-

19

Spee

d &

age

Boy

s & G

irls

VO

2max

= 3

1.02

5 +

3.23

8S -

3.24

8*A

+ 0

.153

6*S*

A

A is

the

age;

S th

e fin

al sp

eed

(S =

8+

0.5

x la

st st

age

com

plet

ed)

Stic

klan

d et

al.

(43)

63

Boy

s

62 G

irls

18-3

8 La

st h

alf-

stag

e co

mpl

eted

& g

ende

r

Boy

s V

O2m

ax =

2.7

5*X

+ 2

8.8

Fe

mal

es

VO

2max

= 2

.85*

X +

25.

1 X

is th

e la

st h

alf-s

tage

com

plet

ed

Fluo

ris e

t al.

(13)

11

0 B

oys

21 +

/- 2.

5 Sp

eed

Boy

s V

O2m

ax =

(S*6

.65-

35.8

)*0.

95 +

0.1

82

S is

the

max

imal

atta

ined

spee

d

Mah

ar e

t al.

(23)

61

Boy

s

74 G

irls

12-1

4 La

ps c

ompl

eted

&

gend

er &

bod

y w

eigh

t

Boy

s & G

irls

VO

2max

= 4

7.43

8 +

(S*0

.242

) + (G

*5.1

34) -

B

M*0

.197

S

is n

umbe

r of l

aps c

ompl

eted

; G is

gen

der (

mal

es=

1,

fem

ale=

0); B

M is

bod

y m

ass (

kg)

TABLE 2. Physical characteristics and 20m shuttle run performance of the study participants

by gender.

All (n = 193) Males (n = 122) Females (n = 71)

Age (yr) 16.1 ± 1.2 16.2 ± 1.3 15.9 ± 1.1

Height (cm) 168.3 ± 9.1 172.5 ± 6.7 161.0 ± 8.2*

Weight (kg) 64.6 ± 13.3 68.5 ± 13.5 58.0 ± 9.8*

Stage 6.5 ± 2.4 8.0 ± 1.7 4.0 ± 1.1*

Speed (km/h) 11.3 ± 1.2 12.0 ± 0.9 10.0 ± 0.5*

Time (min) 6.6 ± 2.4 8.0 ± 1.7 4.1 ± 1.1*

Heart rate (beats/min) 197.7 ± 7.9 198.6 ± 7.9 196.2 ± 7.7

Léger-VO2max (ml/kg/min) 43.0 ± 6.8 47.0 ± 5.0 36.2 ± 2.9*

Measured VO2max (ml/kg/min) 47.7 ± 10.0 53.9 ± 6.2 37.1 ± 5.0*

Data are mean ± SD.

*P < 0.001 from comparisons between sexes.

TABLE 3. Bivariate correlation analysis between measured VO2max (ml/kg/min), age, weigh,

height and speed in males and females.

Age Weight Height Stage†

Males (n = 122)

VO2max r -0.238* -0.517*** -0.160 0.736***

Age r 0.414*** 0.252** 0.057

Weight r 0.550*** -0.195*

Height r -0.070

Females (n = 71)

VO2max r 0.501*** -0.241* 0.219 0.813***

Age r -0.081 0.147 0.418***

Weight r 0.183 -0.118

Height r 0.249*

*P < 0.05; **P < 0.01; ***P < 0.001. † Refers to the last half-stage completed.

TABLE 4. Evaluation of the error of the VO2max measurements obtained from Léger’s

equation and the artificial neural network (ANN)-equation.

Equation

Error measure Léger ANN

∑=

−=N

iii yySSE

1

2)ˆ( 8663.14 1600.91

∑=

−=N

iii yy

NMSE

1

2)ˆ(1 44.89 8.29

MSERMSE = 6.70 2.88

100thdomain wid

% ×=RMSEError 17.13 7.37

)1( 2yyy RSDSEE −= 4.27 2.84

SSE indicates sum of squared errors; MSE: mean of squared errors; RMSE: root mean

squared errors; SEE indicates standard error of estimate. N are the cases available to evaluate

the model where y is the actual output (measure VO2max) and ŷ is the output computed by the

net (ANN-VO2max).

FIGURE 1. Neural network architecture. *Last half-stage completed.

*

Sex

Age

Weight

Height

Stage

Inputs Hidden Output

VO2max

*

Sex

Age

Weight

Height

Stage

Inputs Hidden Output

VO2max

FIGURE 2. Relationship between estimated VO2max from Léger’s equation and measured

VO2max. Crossed line represents the line of equality.

20.0

25.0

30.0

35.0

40.0

45.0

50.0

55.0

60.0

65.0

70.0

20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0 70.0

Estim

ated

Lég

er-V

O2m

ax (m

l/kg/

min

)

Measured VO2max (ml/kg/min)

FIGURE 3. Relationship between estimated VO2max from artificial neural network (ANN)-

equation and measured VO2max. Crossed line represents the line of equality.

20.0

25.0

30.0

35.0

40.0

45.0

50.0

55.0

60.0

65.0

70.0

20.0 25.0 30.0 35.0 40.0 45.0 50.0 55.0 60.0 65.0 70.0

Estim

ated

AN

N-V

O2m

ax (m

l/kg/

min

)

Measured VO2max (ml/kg/min)

FIGURE 4. Bland-Altman plot for the measured VO2max and estimated VO2max from Léger’s

equation. Central line represent the mean difference between equations (4.9 ml/kg/min) and

broken lines represent upper and lower limits of agreement (± 95 Confident Intervals: -4.3 to

14.1 ml/kg/min).

-20.0

-15.0

-10.0

-5.0

0.0

5.0

10.0

15.0

20.0

20 30 40 50 60 70

Diff

eren

ce m

easu

red

VO

2max

& L

éger

-VO

2max

(ml/k

g/m

in)

Mean measured VO2max & Léger-VO2max (ml/kg/min)

FIGURE 5. Bland-Altman plot for the measured VO2max and estimated VO2max from artificial

neural network (ANN)-equation. Central line represent the mean difference between

equations (0.5 ml/kg/min) and broken lines represent upper and lower limits of agreement (±

95 Confident Intervals: -5.1 to 6.1 ml/kg/min).

-20.0

-15.0

-10.0

-5.0

0.0

5.0

10.0

15.0

20.0

20 30 40 50 60 70Diff

eren

ce m

easu

red

VO

2max

& A

NN

-VO

2max

(ml/k

g/m

in)

Mean measured VO2max & ANN-VO2max (ml/kg/min)

HAND SPAN INFLUENCES OPTIMAL GRIP SPAN IN MALE

AND FEMALE TEENAGERS

Jonatan R. Ruiz1,2, Vanesa España-Romero1, Francisco B.

Ortega1,2, Michael Sjöström2, Manuel J. Castillo1, Ángel

Gutiérrez1

J Hand Surg [Am] 2006; 31: 1367-1372


Granada, Spain. 2 Unit for Preventive Nutrition, Department of Biosciences and

Nutrition at NOVUM, Karolinska Institutet, Huddinge, Sweden.

VIII

Hand Span Influences Optimal Grip Spanin Male and Female Teenagers

Jonatan R. Ruiz, BSch, Vanesa España-Romero, BSch,Francisco B. Ortega, BSch, Michael Sjöström, MD, PhD,Manuel J. Castillo, MD, PhD, Angel Gutierrez, MD, PhD

From the Department of Physiology, School of Medicine, University of Granada, Granada, Spain; and theUnit for Preventive Nutrition, Department of Biosciences and Nutrition at NOVUM, Karolinska Institutet,Huddinge, Sweden.

Purpose: To determine if there is an optimal grip span for determining the maximum handgripstrength in male and female teenagers, and if the optimal grip span was related to hand span.If they are related then the second aim was to derive a mathematic equation relating handspan and optimal grip span.Methods: One hundred healthy teenage boys (15.1 � 1.1 y) and 106 girls (15.4 � 1.3 y) wereevaluated (age range, 13–18 y). Each hand was randomly tested on 10 occasions using 5different grip spans, allowing a 1-minute rest between attempts. The hand span was measuredfrom the tip of the thumb to the tip of the small finger with the hand opened as wide aspossible.Results: The results showed that an optimal grip span to determine the maximum handgripstrength was identified for both genders, and the optimal grip span and hand span correlatedin both genders.Conclusions: The results suggest that there is an optimal grip span to which the dynamometershould be adjusted when measuring handgrip strength in teenagers. The optimal grip spanwas influenced by hand span in both genders. For males the optimal grip span can be derivedfrom the equation y � x/7.2 � 3.1 cm, and for females from the equation y � x/4 � 1.1 cm.where y is the optimal grip span and x is the hand-span. These equations may improve thereliability and accuracy of the results and may guide clinicians and researchers in selectingthe optimal grip span on the hand dynamometer when measuring handgrip strength inteenagers. (J Hand Surg 2006;31A:1367–1372. Copyright © 2006 by the American Societyfor Surgery of the Hand.)Key words: Dynamometry, handgrip strength, reliability, standardization, young subjects.

The handgrip strength test is a simple and eco-nomic test that gives practical informationabout muscle, nerve, bone, or joint disor-

ders.1–5 In adults, handgrip strength has been pro-posed as a possible predictor of mortality and theexpectancy of being able to live independently.6,7

The measure of handgrip strength is influenced byseveral factors including age; gender; different angle ofshoulder, elbow, forearm, and wrist8–10; posture9,11;and grip span.9,11–15

Another important factor affecting handgripstrength is hand span.14,15 Several attempts havebeen made to find the optimal grip span that results inmaximum handgrip strength and that increases reli-

able and reproducible handgrip strength in adult andelderly populations. Härkönen et al14 showed thathandgrip strength varied with handgrip position andwas slightly affected by hand span. We have shownthat there is an optimal grip span at which the max-imum handgrip strength is obtained in adults.15

Moreover, the optimal grip span has been shown tobe influenced by individual hand span in adultwomen, but not in men. This can be in relation to thesmaller hand span and/or less grip strength in womencompared with men. Teenagers also present a smallerhand span and less handgrip strength than adults.Handgrip strength is a widely used test in experimen-tal and epidemiologic studies.

The Journal of Hand Surgery 1367

The first aim of the present study was to determineif there is an optimal grip span for determining themaximum handgrip strength in male and female teen-agers, and if that grip span is related to hand span. Ifthese are related than the second aim was to derive amathematic equation relating hand span and optimalgrip span.

Materials and MethodsSubjectsOne hundred boys (15.1 � 1.1 y) and 106 girls (15.4 �1.3 y), with an age range of 13 to 18 years, volun-teered to participate in the study after receiving in-formation about the aim and clinical implications ofthe investigation. The study was conducted in 3schools located in 3 different geographic areas ofSpain. All of the teenagers included in the presentstudy were in good health and free of any lesion orimpairments in the upper limbs. The subjects wereencouraged to do their best when performing thetests. The study was approved by the Review Com-mittee for Research Involving Human Subjects at ourUniversity.

MethodsMeasurement of hand span. Hand span wasmeasured in both hands from the tip of the thumb tothe tip of the small finger with the hand opened aswide as possible (Fig. 1). The precision of the mea-sure was 0.5 cm, but the results of the hand spanmeasurement were rounded to the nearest wholecentimeter.

Measurement of handgrip strength. Handgripstrength was measured using a digital dynamometer

(T.K.K. 5101 Grip-D; Takey, Tokyo, Japan), and thescores were recorded in kilograms. The reported pre-cision of the dynamometer was 0.1 kg. When per-forming the measurement, subjects were instructed tomaintain the standard bipedal position during theentire test with the arm in complete extension and notto touch any part of the body with the dynamometerexcept the hand being measured. Each subject per-formed (alternately with both hands) the test twiceusing different grip spans in random order, allowinga 1-minute rest between the measurements.11 Foreach measure, the hand to be tested first was chosenrandomly. The grip spans used were 4.5, 5.0, 5.5, 6.0,6.5, and 7.0 cm. If the hand span was less than 20 cmthen the highest grip span was rejected; if the handspan was more than 20 cm then the lowest grip spanwas rejected. For each hand the best result for eachgrip span was retained. For the hand dynamometer(Jamar; Fit Systems Inc., Calgary, Canada) the gripspan equivalence for the different positions are asfollows: position 1, 3.5 cm; position 2, 4.8 cm; po-sition 3, 6.0 cm; position 4, 7.3 cm; and position 5,8.6 cm.

Determination of optimal grip span. The optimalgrip span is the grip span at which the maximumhandgrip strength is obtained. To determine the in-dividual optimal grip span for each hand of eachindividual we first established the kind of associationrelating grip span and handgrip strength (ie, the re-sults of handgrip strength obtained at the different gripspans). For that purpose, statistical software (SPSSv.14.0; SPSS Inc., Chicago, IL) was used. The asso-ciation could be lineal, logarithmic, potential, expo-nential, or polynomial. In all subjects (except for 6)the association was statistically significant. All func-tions were considered, and the most relevant one wasretained. The mathematic function of the relation wasindividually determined through the least-square fitand graphically represented (Fig. 2). In 190 of thepatients it was quadratic and parabolic (correspond-ing to a second-degree polynomial equation). Once wedefined the equation, the optimal grip span was calcu-lated as x/f’(x) � 0, where x equals the optimal gripspan (cm) and f(x) equals the handgrip strength (kg).In graphic terms, this corresponded to the maximumof the curves (Fig. 2). For nonpolynomial equations(n � 16), the optimal grip span was graphicallydetermined and this corresponded to one of the ex-treme grip spans used for that particular subject. Inthose subjects in whom there was no associationFigure 1. Measure of hand span (0.5-cm precision).

1368 The Journal of Hand Surgery / Vol. 31A No. 8 October 2006

between handgrip strength and grip span (n � 6), theaverage of the chosen grip spans was retained.

Determination of the optimal grip span for a givenhand span. By using statistical software (SPSSpackage v.14.0), we studied whether optimal grip spanswere significantly related to hand spans (p � .05). Incase of a significant relationship, we used the least-square fit to calculate the mathematic function relatingboth variables. This equation allows the establishmentof the optimal grip span for a given hand span. In caseof a nonsignificant relationship, the conclusion is thatoptimal grip spans are not related to hand spans.

Usefulness and reliability of the optimal gripspan. To confirm the usefulness of using the opti-mal grip span when measuring handgrip strength, anadditional group of 21 teenagers (13 males, 8 fe-males) ages 14 to 17 years volunteered to perform thehandgrip strength test at 3 grip spans: optimal gripspan, 1 cm below the optimal grip span, and 1 cmabove the optimal grip span. Each subject performed(alternately with both hands) the test twice using dif-ferent grip spans in a random order, allowing a1-minute rest between the measurements.11 For eachmeasure, the hand to be tested first was chosen ran-domly. For each hand the best result at each grip spanwas retained.

To confirm the reliability of measurements ofhandgrip strength at the optimal grip span, 17 (13males, 4 females) of the previous 21 teenagers lessthan 18 years of age performed the test at the optimalgrip span 3 hours later. The subjects were advised notto perform strenuous exercise during the 3 hourspreceding the second test.

Statistical AnalysisThe normality of the distribution of the measuredvariables was ascertained by the Shapiro-Wilk test.The hand span, handgrip strength, and the optimalgrip span obtained for each hand span was comparedby 1-way analysis of variance (ANOVA). Bivariatecorrelation analysis was performed to examine therelationship between optimal grip span and handspan for each hand and gender. In case of an associ-ation, the mathematic function defining the associa-tion was calculated through the least-square fit.

For confirming the usefulness of measuring hand-grip strength at the optimal grip span, 1 cm below theoptimal grip span, and 1 cm above the optimal gripspan, a 1-way ANOVA was used. The reliability coef-ficient of handgrip strength measured at the optimalgrip span on 2 different occasions was calculated;values were compared through 1-way ANOVA andcorrelated through parametric bivariate correlationanalysis. The � error was fixed at .05.

ResultsAll subjects completed the tests satisfactorily. Themean � SD measured hand span was 21.0 � 1.3 cmfor males (n � 100) and 18.7 � 1.1 cm for females(n � 106) (p � .001). Males obtained higher valuesof handgrip strength at each grip span than females(p � .01) (data not shown). In both genders, and forboth hands, an optimal grip span was obtained. Theoptimal grip span for each hand span for males andfemales is presented in Tables 1 and 2, respectively.No significant differences were obtained betweenboth hands for each hand span (p � .70). Because theoptimal grip span was not different between the rightand left hands, the mean value was retained and usedfor subsequent analysis.

Table 1. Optimal Grip Span Determined inFemales (n � 106) for Each Hand Span

Hand Span,cm

OptimalGrip Spanfor RightHand, cm

OptimalGrip Span

for LeftHand, cm

Optimal GripSpan, cm*

16 5.0 � 0.7 4.9 � 0.5 5.017 5.6 � 0.7 5.6 � 0.6 5.618 5.5 � 0.7 5.5 � 0.6 5.519 5.8 � 0.6 5.8 � 0.5 5.820 5.8 � 0.5 6.4 � 0.6 6.1

The precision of the hand-span measurement was 0.5 cm andwas rounded to the nearest whole centimeter. No significantdifferences were obtained between both hands for each handspan (p � .70).

*Optimal grip span obtained from the mean of the right- andleft-hand optimal grip spans.

Figure 2. Association of handgrip strength and grip span in 1subject. The maximum of the second-degree polynomialregression equation relating handgrip strength and grip span[f’(x)] was the optimal grip span for each hand of eachindividual. f(x) � �5.7143x2 � 63.857x � 149.44; f’(x) ��11.4286 � 63.857; x | [f’(x) � 0] � 5.6 cm.

Ruiz et al / Handgrip Strength in Adolescents 1369

In teenagers, hand span and optimal grip spanshowed a significant linear association (y � 0.16x �2.66; r � .92, p � .001) where x is the hand span, andy is the optimal grip span at which the dynamometershould be adjusted before the test. The equation re-lating grip span as a function of hand span in males isformulated as y � 0.1386x � 3.101 (r � .92, p � .01).A simplification of this algorithm would be the fol-lowing: y � x/7.2 � 3.1 (Fig. 3). The equationrelating grip span as a function of hand span infemales is formulated as y � 0.25x � 1.09 (r � .93,p � .02). A simplification of this algorithm would bethe following: y � x/4 � 1.1 (Fig. 3). Table 3 showsthe optimal grip span calculated from the equationsprovided, for each hand span in males and females.

The handgrip strength obtained at the optimal gripspan was significantly higher (p � 0.006) than thestrength obtained when the grip was set 1 cm below or1 cm above the optimal grip span, in both hands andgenders (Fig. 4).

Seventeen adolescents (13 males, 4 females) from

the previous 21 repeated the test 3 hours later at theoptimal grip span. The results showed a reliabilitycoefficient of 0.98 and 0.96 for the right and lefthands, respectively. Moreover, the 1-way ANOVAdid not show a statistical difference between the testand retest results (p � .45 and .53 for the right andleft hands, respectively). A significant correlationbetween the test and retest results was obtained forright (r � .96, p � .001) and left (r � .92, p � .001)hands at the optimal grip span.

DiscussionThis study suggests that there is an optimal grip spanto which the standard dynamometer should be ad-

Table 2. Optimal Grip Span Determined inMales (n � 100) for Each Hand Span

Hand Span,cm

OptimalGrip Spanfor RightHand, cm

OptimalGrip Span

for LeftHand, cm

Optimal GripSpan, cm*

18 5.3 � 0.7 5.6 � 0.9 5.519 5.9 � 0.5 5.7 � 0.9 5.820 6.1 � 0.6 6.0 � 0.6 6.121 6.0 � 0.6 6.0 � 0.7 6.022 6.0 � 0.6 6.2 � 0.7 6.123 6.2 � 0.8 6.3 � 0.6 6.3

The precision of the hand-span measurement was 0.5 cm andwas rounded to the nearest whole centimeter.

No significant differences were obtained between both handsfor each hand span (p � .70).

*Optimal grip span obtained from the mean of the right- andleft-hand optimal grip spans.

FemalesMales

y = 0.1386x + 3.101y = x/7.2 + 3.1

5.05.25.45.65.86.06.26.46.66.87.0

17 18 19 20 21 22 23 24

Hand-span (cm)

Op

tim

al g

rip

sp

an (

cm)

r = .92P = .01

y = 0.25x + 1.09y = x/4 + 1.1

4.8

5.0

5.2

5.4

5.6

5.8

6.0

6.2

15 16 17 18 19 20 21

Hand-span (cm)

Op

tim

al g

rip

sp

an (

cm)

r = .93P = .02

A B

Figure 3. Association between hand span and optimal grip span in (A) males (n � 100) and (B) females (n � 106). (A) y �0.1386x � 3.101; y � x/7.2 � 3.1; r � 0.92; p � .01. (B) y � 0.25x � 1.09; y � x/4 � 1.1; r � 0.93; p � .02.

Table 3. Optimal Grip Span for Each Hand SpanCalculated From the Equations Provided

Hand Span,cm

OptimalMale andFemale

Grip Span, cm

OptimalMale GripSpan, cm

OptimalFemale Grip

Span, cm

16.0 5.2 5.3 5.116.5 5.3 5.4 5.217.0 5.4 5.5 5.417.5 5.5 5.5 5.518.0 5.5 5.6 5.618.5 5.6 5.7 5.719.0 5.7 5.7 5.919.5 5.8 5.8 6.020.0 5.9 5.9 6.120.5 5.9 5.9 6.221.0 6.0 6.0 6.421.5 6.1 6.1 6.522.0 6.2 6.1 6.622.5 6.3 6.2 6.723.0 6.3 6.3 6.9

For males and females: y � 0.16x � 2.66 (r � .92, p � .001);males: y � x/7.2 � 3.1 (r � .92, p � .01); females: y � x/4 � 1.1(r � .93, p � .02), where x is the hand span (maximal widthbetween the thumb and small finger, with 0.5-cm precision), andy is the optimal grip span in cm.


justed when measuring handgrip strength in bothmales and females ages 13 to 18 years. In bothgenders the optimal grip span is influenced by handspan, which implies the need for adjustment of thegrip span of the dynamometer to the hand span. Forthat purpose gender-specific equations are proposed,and are valid for both hands. Handgrip strength is awidely used test in experimental and epidemiologicstudies in young people.

We have previously shown similar results in adultmen and women.12 In women the optimal grip spanwas influenced by hand span, and an equation tocalculate the optimal grip span from the measure ofthe hand span was proposed (y � x/5 � 1.5). In menthere was an optimal grip span for determining themaximum handgrip strength, but that optimal gripspan was not hand-span dependent; therefore a fixedoptimal grip span was proposed (5.5 cm). Teenagershave smaller hand spans and less handgrip strengthcompared with adults. Because of these differencesone would expect that teenagers may need a differentoptimal grip span when measuring handgrip strengthcompared with adults. In the present study, the opti-mal grip span was influenced by hand span in bothmale and female teenagers, similar to what we foundpreviously in adult women, but not in adult men.Adult men, usually already part of the workforce(mostly manual workers), might compensate for thehand-span effect with higher muscle mass and mus-cle strength in their forearm. This could partiallyexplain the lack of association between the hand spanand the optimal grip span in adult men.

Other studies also have shown a specific grip spanat which the maximum handgrip strength is ob-tained.11–13,16,17 Middle grip spans seem to favorgreater forces than smaller or larger grips.16 Oh and

Radwin17 reported that hand span affected maximaland submaximal handgrip strengths. They found thathand span affected grip strength, grip force, andexertion level. In another study,13 the optimal gripspan was suggested to be 5.0 to 6.0 cm for womenand 5.5 to 6.5 cm for men. Similar values have beenfound recently in a larger study11 in which the sub-jects performed the handgrip test at 3 different gripspans: one grip span, called the standard grip span,was calculated from the half distance between theindex fingertip and the metacarpophalangeal jointflexion crease at the base of the thumb (men, 5.8 cm;women, 5.4 cm), the other grip spans were at �10%and �10% of the standard grip span. It was con-cluded that the grip span that achieves maximumhandgrip strength is somewhere between the standardgrip span and a 10% increase of that distance. Theage and the number of participants in the earlier-mentioned studies make comparisons difficult.

Different measures of handgrip strength are currentlyused worldwide. There are some international physicalfitness test batteries specifically designed for the youngpopulation that include a handgrip strength test (eg,EUROFIT test battery18). From a public health perspec-tive it is important to standardize the procedure andincrease the reliability because otherwise the measure-ment error may be too large to detect actual changes instrength; however, different kinds of dynamometersand postures might change the results. We do not knowwhether these findings can be directly transferred tomeasurements with other dynamometers.

Received for publication April 17, 2006; accepted in revised form June 26,2006.

The present article is published on behalf of the HELENA (HealthyLifestyle in Europe by Nutrition in Adolescence) Study group (http://www.helenastudy.com/list.php).

32

34

36

38

40

42

44

Han

dgr

ip s

tren

gth

(k

g)

Right hand

Left hand

1 cm below Optimal grip span 1 cm above

*

†

18

20

22

24

26

28

30

Han

dgr

ip s

tren

gth

(k

g)

Right hand

Left hand

1 cm below Optimal grip span 1 cm above

*

†

Males Females

A B

Figure 4. Handgrip strength measured for the right and left hands at the optimal grip span, 1 cm below the optimal grip span,and 1 cm above the optimal grip span in (A) males (n � 13) and (B) females (n � 8) (age range, 14–17 y). The values are mean �standard error of the mean. *p � .005 compared with 1 cm below and 1 cm above the optimal grip span. †p � .006 comparedwith 1 cm below and 1 cm above the optimal grip span. (A) �, right hand; □, left hand; (B) ●, right hand; Œ, left hand.

Ruiz et al / Handgrip Strength in Adolescents 1371

No benefits in any form have been received or will be received froma commercial party related directly or indirectly to the subject of thisarticle.

The HELENA study was supported by the European Community SixthRTD Framework Programme (Contract FOOD-CT-2005-007034). Alsosupported by a grant from the Ministerio de Educación y Ciencia deEspaña (AP2003-2128, AP2004-2745 to J.R.R. and F.B.O.).

The contents of this article reflect only the authors’ views and theEuropean Community is not liable for any use that may be made of theinformation contained therein.

Corresponding author: Jonatan R. Ruiz, BSch, Department of Physi-ology, School of Medicine, University of Granada, 18071 Granada,Spain; e-mail: [email protected].

Copyright © 2006 by the American Society for Surgery of the Hand0363-5023/06/31A08-0018$32.00/0doi:10.1016/j.jhsa.2006.06.014

References1. Schreuders TAR, Roebroeck M, Van der Kar JM, Soeters

JNM, Hovius SER, Stam HJ. Strength of the intrinsic mus-cles of the hand measured with a hand-held dynamometer:reliability in patients with ulnar and median nerve paralysis.J Hand Surg 2000;25B:560–565.

2. Ozgocmen S, Karaoglan B, Cimen OB, Yorgancioglu ZR.Relation between grip strength and hand bone mineral den-sity in healthy women aged 30–70. Singapore Med J 2000;41:268–270.

3. Wessel J, Kaup C, Fan J, Ehalt R, Ellsworth J, Speer C, et al.Isometric strength measurements in children with arthritis:reliability and relation to function. Arthritis Care Res 1999;12:238–246.

4. Merkies IS, Schmitz PI, Samijn JP, Meche FG, Toyka KV,van Doorn PA. Assessing grip strength in healthy individu-als and patients with immune-mediated polyneuropathies.Muscle Nerve 2000;23:1393–1401.

5. Di Monaco M, Di Monaco R, Manca M, Cavanna A. Hand-grip strength is an independent predictor of distal radiusbone mineral density in postmenopausal women. Clin Rheu-matol 2000;19:473–476.

6. Metter EJ, Talbot LA, Schrager M, Conwit R. Skeletalmuscle strength as a predictor of all-cause mortality in

healthy men. J Gerontol A Biol Sci Med Sci 2002;57:B359–B365.

7. Seguin R, Nelson ME. The benefits of strength trainingfor older adults. Am J Prev Med 2003;25(suppl 2):S141–S149.

8. Su C-Y, Lin JH, Chien TH, Cheng KF, Sung YT. Gripstrength in different positions of elbow and shoulder. ArchPhys Med Rehabil 1994;75:812–815.

9. Mathiowetz V, Rennells C, Donahoe L. Effect of elbowposition on grip and key pinch strength. J Hand Surg 1985;10A:694–697.

10. Richards LG, Olson B, Palmiter-Thomas P. How forearmposition affects grip strength. Am J Occup Ther 1996;50:133–138.

11. Watanabe T, Owashi K, Kanauchi Y, Mura N, Takahara M,Ogino T. The short-term reliability of grip strength measure-ment and the effects of posture and grip span. J Hand Surg2005;30A:603–609.

12. Firrell JC, Crain GM. Which setting of the dynamometerprovides maximal grip strength? J Hand Surg 1996;21A:397–401.

13. Fransson C, Winkel J. Hand strength: the influence of gripspan and grip type. Ergonomics 1991;34:881–892.

14. Härkönen R, Piirtomaa M, Alaranta H. Grip strength andhand position of the dynamometer in 204 Finnish adults.J Hand Surg 1993;18B:129–132.

15. Ruiz-Ruiz J, Mesa JL, Gutiérrez A, Castillo MJ. Hand sizeinfluences optimal grip span in women but not in men.J Hand Surg 2002;27A:897–901.

16. Blackwell JR, Kornatz KW, Heath EM. Effect of grip spanon maximal grip force and fatigue of flexor digitorum su-perficialis. Appl Ergon 1999;30:401–405.

17. Oh S, Radwin RG. Pistol grip power tool handle and triggersize effects on grip exertions and operator preference. HumFactors 1993;35:551–569.

18. Committee of Experts on Sports Research EUROFIT. Hand-book for the EUROFIT Tests of Physical Fitness. Strasburg,GE: Council of Europe, 1993:19–37.


A MEDITERRANEAN DIET IS NOT ENOUGH FOR HEALTH:

PHYSICAL FITNESS IS AN IMPORTANT ADDITIONAL

CONTRIBUTOR TO HEALTH FOR THE ADULTS OF

TOMORROW

Manuel J. Castillo, Jonatan R. Ruiz, Francisco B. Ortega, Angel

Gutierrez

World Rev Nutr Diet 2007; 97: 114-138

School of Medicine, University of Granada and Sotogrande Health Experience, Spain

IX

Proo

f

Simopoulos AP, Visioli F (eds): More on Mediterranean Diets. World Rev Nutr Diet. Basel, Karger, 2007, vol 97, pp 114–138

A Mediterranean Diet Is Not Enough forHealth: Physical Fitness Is an ImportantAdditional Contributor to Health for theAdults of Tomorrow

Manuel J. Castillo-Garzón, Jonatan R. Ruiz, Francisco B. Ortega, Angel Gutierrez-Sainz

School of Medicine, University of Granada and Sotogrande Health Experience,Granada, Spain

Is It Only Diet?

Cardiovascular diseases are the major cause of death in Western Societies.Nevertheless, important differences exist among different populations andregions. In Europe, for instance, large differences exist in mortality from coro-nary heart disease and stroke. These diseases show a clear West-East and South-North gradient with high rates in Eastern and Northern Europe and lower ratesin most Mediterranean countries [1]. Interestingly, large regional differencesin ischemic heart disease and prevalence of cardiovascular risk factorsoccur within the same country and even within the same region. These differ-ences are present both in countries with high and low incidence of cardiovascu-lar disease [2].

Classical risk factors for cardiovascular disease include age, sex, hyperten-sion, smoking, diabetes, elevated plasma low-density lipoprotein (LDL)-cho-lesterol, and low high-density lipoprotein (HDL)-cholesterol, lack of exerciseand increased body fat. Nevertheless, the contribution of changes in these fac-tors to trends in coronary event rates can only explain half of the cases [3].Emerging independent risk factors include abdominal adiposity, high plasmalevels of triglycerides, lipoprotein(a), modified LDL-cholesterol particles,homocysteine, several markers of inflammation, and thrombotic risk factors. Itis quite possible that even taking all these factors into account, differences incoronary heart disease rates could not be fully explained.

WRN97114.qxd 3/11/06 4:37 PM Page 114

Proo

f

�� 115

Growing evidence demonstrates that the Mediterranean life-style is bene-ficial to health. The evidence is stronger for coronary heart disease, but it alsoapplies to stroke and some forms of cancer [4–6]. Diet is one outstanding com-ponent of the Mediterranean life-style. Large life-style and dietary variationsoccur in different regions and countries. In many of them, a progressive depar-ture from the traditional Mediterranean life-style and diet is being observed [7].In this departure, more affluent economies and younger subjects are, probably,more easily influenced.

In addition to diet, a sedentary life-style is a major risk factor for noncom-municable diseases (e.g. coronary heart disease, stroke, obesity, hypertension,type 2 diabetes, allergies and several types of cancers) and is close to overtakingtobacco as the leading cause of preventable death [8]. The protective effect ofregular physical activity on the above mentioned diseases has been widelyreported in young people, in adults and in the elderly. It is now well known thatregular participation in moderate and vigorous levels of exercise can lead tomany health benefits (table 1).

The Spanish-Mediterranean Life-Style (and Diet)

The Spanish-Mediterranean life-style (and diet) is that usually followed bythe inhabitants of Spain. Nevertheless, geographical, economic and social differ-ences result in many different dietary practices and physical activity patterns.This, obviously, precludes a single definition of the Spanish-Mediterranean life-style. Nonetheless, regarding diet, there is a dietary pattern that is common in thedifferent diets in the country. This traditional dietary pattern is composed of acluster of basic foods that have been easily available in the region during

Table 1. Beneficial effects on health of practising regular physical exercise

Reduction in the risk of developing ischemic heart disease and other cardiovascular diseasesReduction in the risk of developing obesity and diabetesReduction in the risk of developing (and control of) high blood pressure and dyslipidemiaReduction in the risk of developing breast and colon cancerHelps in the control of body weight and improves ‘body image’Tonifies muscles and preserves or increases muscular massStrengthens bones and jointsIncreases coordination and neuromotor responses; reduces the risk of fallsImproves immune system activityReduces depression and anxietyPromotes wellbeing and social integration

WRN97114.qxd 3/11/06 4:37 PM Page 115

Proo

f

Castillo-Garzón/Ruiz/Ortega/Gutierrez-Sainz 116

centuries. This determines a diet that is high in fruits, green and root vegetables,bread, other forms of cereals, beans, nuts and seeds of different types. A com-mon and outstanding characteristic of the Spanish Mediterranean diet is the useof olive oil which represents the more important source of fat in the diet. Animalproducts intake includes eggs, dairy products and poultry. There are significantvariations in the intakes of fish, red meat (pork, beef, lamb) and meat derivedproducts. Red wine and beer have been traditionally consumed. A main differ-ence between the Spanish-Mediterranean diet and other Mediterranean diets isthe lower intake of pasta and potatoes, and the higher intake of bread, legumesand fish [9, 10]. Many of these components may have an effect on cardiovascu-lar risk factors, particularly by influencing the plasma lipid profile.

One specific characteristic of the traditional Spanish-Mediterranean life-style is the time spent outdoor which is favored by the favorable weather condi-tions. This may determine higher levels of physical activity. In Spain, thetimetable for meals is different from other countries. The main meal of the dayis usually taken in early afternoon (around 2–3 p.m.) and the dinner is late in theevening (around 10 p.m.) and rather light. There is a widely spread culture ofeating outside the home in an informal way and usually standing up. It is thetypical ‘tapas’ eating. These ‘tapas’ are taken between meals and occasionallyrepresent an alternative to a more formal meal.

Diet and Physical Activity Interaction

Diet and physical activity interact in the development and prevention ofischemic heart disease and several other health conditions. Both factors affectthe plasma lipid profile and body composition, and probably influence otherrisk factors. In fact, a physiological means of influencing diet-induced modifi-cations of the plasma lipid profile and body fat content is physical activity [11].Physical activity favorably influences all three components of the atherogeniclipoprotein phenotype: the HDL-cholesterol concentration may increase, LDL-cholesterol may decrease, and serum triglycerides can also be reduced [12, 13].In addition, physical activity precludes body fat accumulation. Complex inter-actions between diet, physical activity, life-style, lipoprotein metabolism andother factors determine the development of atherosclerosis and its complica-tions. These interactions may start early in life. In this way, adolescence is acritical period because it is at that time when the individual takes control ofhis/her own life-style and diet.

We have studied a representative sample of Spanish adolescents aged13–18.5 years participating in the ‘Alimentación y Valoración del EstadoNutricional de los Adolescentes’ (AVENA) study (www.estudioavena.com). The

WRN97114.qxd 3/11/06 4:37 PM Page 116

Proo

f

�� 117

AVENA study is a population-based cross-sectional survey conducted in fivedifferent geographic areas of Spain (Madrid, Murcia, Granada, Santander andZaragoza), addressing genetic and environmental factors in relation to metabolictraits during adolescence [14]. Some interesting data regarding cardiovascularrisk factors are being obtained from this study. Interestingly, we have observed ahigh prevalence of an unfavorable plasma lipid profile, both in boys (fig. 1) andgirls (fig. 2) [15]. Similarly, it is well known the high prevalence of obesity inMediterranean children and adolescents (fig. 3) [16]. These results underline theimportance of implementing effective measures for preventing the deleteriousconsequences that these conditions are going to have in the health of tomorrow’s

25456585

105125145165185205225245

12 13 14 15 16 17 18 19

TG (m

g/d

l)

25

35

45

55

65

75

85

95

12 13 14 15 16 17 18 19

HD

Lc (m

g/d

l)

30405060708090

100110120130140150

12 13 14 15 16 17 18 19

LDLc

(mg/

dl)

100

125

150

175

200

225

250

12 13 14 15 16 17 18 19

TC (m

g/d

l)

4%

24%

5%

13%

24%

6%

5%

20%

Age (years)

Males

Age (years)

Age (years) Age (years)

Fig. 1. Serum levels of total cholesterol (TC), low-density lipoprotein cholesterol(LDLc), high-density lipoprotein cholesterol (HDLc) and triglycerides (TG) in male Spanishadolescents. Solid lines represent the limit level considered as healthy. Broken lines representthe limit level considered as unhealthy. Subjects between both lines can be considered asborderline.

WRN97114.qxd 3/11/06 4:37 PM Page 117

Proo

f


adults. One positive measure is the return to the traditional Spanish-Mediterraneandiet; the other is to increase the levels of physical activity. In other words, thereturn to the traditional Spanish-Mediterranean life-style.

Physical Activity, Physical Exercise and Physical FitnessRegular physical activity stimulates functional adaptation of all tissues and

organs in the body, thereby also making them less vulnerable to lifestyle-relateddegenerative and chronic diseases. Physical activity refers to any body move-ment produced by muscle action that increases energy expenditure. Physical exer-cise refers to planned, structured, repetitive and purposeful physical activity.

25456585

105125145165185205225245

12 13 14 15 16 17 18 19

TG (m

g/d

l)

25

35

45

55

65

75

85

95

12 13 14 15 16 17 18 19

HD

Lc (m

g/d

l)

30405060708090

100110120130140150

12 13 14 15 16 17 18 19

LDLc

(mg/

dl)

100

125

150

175

200

225

250

12 13 14 15 16 17 18 19

TC (m

g/d

l)

Age (years)

Females

Age (years)

Age (years) Age (years)

11%

34%

7%

21%

21%

2%

4%

9%

Fig. 2. Serum levels of total cholesterol (TC), low-density lipoprotein cholesterol(LDLc), high-density lipoprotein cholesterol (HDLc) and triglycerides (TG) in femaleSpanish adolescents. Solid lines represent the limit level considered as healthy. Broken linesrepresent the limit levels considered as unhealthy. Subjects between both lines can be con-sidered as borderline.

WRN97114.qxd 3/11/06 4:37 PM Page 118

Proo

f

�� 119

Physical fitness is the capacity to perform physical exercise. Physical fitnessmakes reference to the full range of physical qualities, e.g. aerobic capacity,muscle strength, speed, agility, coordination and flexibility. It can be under-stood as an integrated measurement of most, if not all, the body structures andfunctions (skeletomuscular, cardiorespiratory, hematocirculatory, psychoneuro-logical and endocrine-metabolic) involved in the performance of physical activ-ity and/or physical exercise [17]. Thus, being physically fit implies that theresponse of these functions and structures will be adequate. A person cannot bemore physically fit than that allowed by the function or structure in lowest con-dition. Health-oriented physical fitness includes those components of physicalfitness more associated with aspects of good health and/or disease prevention[17].

Physical Fitness as a Health Determinant Aerobic capacity or cardiorespiratory fitness is one of the key compo-

nents of physical fitness. Maximum aerobic capacity is expressed in terms ofmaximum oxygen consumption (VO2max). The VO2max can be expressedwith respect to subject weight (ml/kg/min), in absolute terms (l/min) or inmetabolic equivalents (METs). One MET is the energy expenditure at rest(�3.5 ml/kg/min). Thus, if a subject has a VO2max of 42 ml/kg/min, he/she also

2015

12

18

1922

16

18

18

1712

26

3617

31

3533 27

18

10

34

Fig. 3. Prevalence (%) of children (7–10 years old) with overweight in Europe [16].

WRN97114.qxd 3/11/06 4:37 PM Page 119

Proo

f


has an energy expenditure of 12 METS (i.e. he/she is able to increase his/herresting energy expenditure 12-fold).

A number of important prospective studies have shown that VO2max is themost important predictor of all-cause mortality, and in particular of cardiovas-cular death. This is true for both healthy persons and for people with cardiovas-cular disease [18], and for both men [19–21] and women [22, 23] of differentages [24]. An almost linear reduction in mortality is seen as the cardiorespira-tory fitness increases [23, 24]. For each increase of 1 MET, there is a 12%increase in the life expectancy of men [24] and a 17% increase in women [22].This is even more evident if cardiovascular mortality is considered alone, andagain is true for both men [18, 21] and women [22, 23]. An inverse relationshiphas also been found between cardiorespiratory fitness and mortality due to can-cer independently of age, alcohol intake, diabetes mellitus and tobacco [25–28].Similarly, it has been shown that VO2max is associated with insulin sensitivity[29]; low VO2max levels are also associated with metabolic syndrome (abdomi-nal obesity, glucose intolerance, type II diabetes, hypertension, hyperlipidemiaand insulin resistance) [30, 31]. High levels of cardiorespiratory fitness reducethe neuronal losses associated with aging [32] and protects against cognitivedysfunction [33].

Handgrip strength, assessed by the manual dynamometer test, is currentlyconsidered to be a reliable marker of health and well-being and a potent predic-tor of mortality and the expectancy of being able to live independently [34, 35].Efforts are made to reduce the errors associated with its measurement in ado-lescents [36] and adults [37].

Physical Fitness and Cardiovascular Risk Factors in MediterraneanAdolescents

Cardiorespiratory Fitness and Traditional Cardiovascular Risk FactorsCardiorespiratory fitness is a direct marker of physiological status and

reflects the overall capacity of the cardiovascular and respiratory systems.Results from several cross-sectional studies have clearly shown strong negativeassociations between cardiorespiratory fitness and cardiovascular risk factorsnot only in adults but also in children and adolescents (table 2). In addition,results from prospective studies suggest that high cardiorespiratory fitness dur-ing childhood seems to provide more health protection in adulthood.

Associations between increased cardiorespiratory fitness and several car-diovascular risk factors have been repeatedly found. As it is known, elevatedlevel of triglycerides is strongly associated with an increased risk of coronaryartery disease. In Spanish adolescents (aged 13–18.5 years) it was found a

WRN97114.qxd 3/11/06 4:37 PM Page 120

Proo

f

�� 121

negative correlation between cardiorespiratory fitness and triglycerides, espe-cially in males (fig. 4). In females, a trend toward lower levels of triglycerideswith increasing fitness was also observed. These findings concur with theresults obtained in children and adolescents from other European countries(table 2). Indeed, a negative correlation between cardiorespiratory fitness andtriglycerides has been found in Danish, Swedish and Estonian children from theEuropean Youth Heart Study, which is also in agreement with findings fromtheir American peers (table 2).

Similar associations were also observed between cardiorespiratory fitnessand LDL-cholesterol. There was a trend indicating lower levels of LDL-choles-terol with higher levels of fitness in both male and females (fig. 5). These find-ings are noteworthy since it is known that LDL-cholesterol and their oxidizedderivatives initiate and promote the atherosclerotic process, leading to thedevelopment of coronary artery disease.

Plasma HDL-cholesterol has anti-atherogenic proprieties with its concen-tration inversely related to risk of coronary artery disease. It is estimated thatfor every 1 mg/dl (0.026 mmol/l) increase in HDL-cholesterol, the risk for acoronary heart disease event is reduced by 2% in men and at least 3% inwomen. Cardiorespiratory fitness has been shown to be negatively correlatedwith HDL-cholesterol in children, adolescents and adult population. Figure 6clearly shows the associations between fitness and HDL-cholesterol. This isalso the case for fitness and apolipoprotein (apo) A-1 (fig. 7). Apo A-1 is themost abundant protein of HDL-cholesterol. An increase in the apo A-1 can leadto an increase of HDL-cholesterol. Alternatively, increased catabolism orremoval of apo A-1 will lead to a reduction in plasma HDL-cholesterol levels.

A more favorable metabolic profile (computed with age and gender spe-cific standardized values of triglycerides, LDL-cholesterol, HDL-cholesteroland fasting glycemia) with increased levels of cardiorespiratory fitness has alsobeen shown in Spanish adolescents [13]. Figure 8 shows the association of car-diorespiratory fitness and metabolic profile in non-overweight and overweightadolescents. These results suggest that both fitness and weight management arenecessary for the prevention of lipid-related cardiovascular risk in adolescents.In fact, the odds ratio for having an unfavorable lipid profile is increased in sub-jects with low cardiorespiratory fitness even after adjusting for age and waistcircumference (fig. 9).

Cardiorespiratory Fitness and Emerging Cardiovascular Risk FactorsCardiorespiratory fitness has also been associated with recently recog-

nized cardiovascular risk factors such as low grade inflammation markers (e.g.C-reactive protein, fibrinogen, ceruloplasmin, complement factor C3 and C4) andhomocysteine. Findings from the AVENA study suggest that cardiorespiratory

WRN97114.qxd 3/11/06 4:37 PM Page 121

Proo

f


Tabl

e 2.

Sum

mar

y of

rec

ent s

tudi

es e

xam

inin

g th

e as

soci

atio

ns b

etw

een

card

iore

spir

ator

y fi

tnes

s an

d he

alth

-rel

ated

var

iabl

es in

chi

ldre

nan

d ad

oles

cent

s

Stud

yTy

pe o

f Su

bjec

tsA

ge

Out

com

est

udy

Gut

in e

t al.

[38]

cros

s-bo

ys�

116

14–1

8 ye

ars

boys

and

gir

lsse

ctio

nal

girl

s�

166

CR

F w

as in

vers

ely

asso

ciat

ed w

ith in

sulin

co

ncen

trat

ions

, and

the

adve

rse

impa

ct o

f lo

w

CR

F w

as g

reat

er in

boy

s th

an in

gir

ls

Bra

ge e

t al.

[39]

cros

s-bo

ys�

279

8–

10 y

ears

boys

and

gir

lsse

ctio

nal

girl

s�

380

CR

F w

as in

vers

ely

asso

ciat

ed w

ith in

sulin

, TG

, sy

stol

ic B

P, a

nd s

kinf

old

thic

knes

ses

(p�

0.03

3).

CR

F w

as in

vers

ely

asso

ciat

ed w

ith m

etab

olic

sy

ndro

me

Z s

core

(p

�0.

031)

.C

RF

was

pos

itive

ly a

ssoc

iate

d w

ith H

DL

c (p

�0.

002)

Ree

d et

al.

[40]

cros

s-bo

ys�

559–

11 y

ears

boys

and

gir

lsse

ctio

nal

girl

s�

44C

RF

acco

unte

d fo

r 37

% o

f th

e va

rian

ce in

larg

e ar

tery

com

plia

nce.

Hig

hest

CR

F qu

artil

e ha

d gr

eate

r co

mpl

ianc

e th

an c

hild

ren

in th

e tw

o lo

wes

t

CR

F qu

artil

es, b

y as

muc

h as

34%

Eis

enm

ann

cros

s-B

oys

�41

69–

18 y

ears

CR

F an

d B

MI

show

ed a

n in

depe

nden

t ass

ocia

tion

et a

l. [4

1]se

ctio

nal

Gir

ls�

345

with

car

diov

ascu

lar

risk

fac

tors

Gut

in e

t al.

[42]

cros

s-bo

ys�

187

14–1

8 ye

ars

high

er C

RF

and

low

er f

atne

ss w

ere

asso

ciat

ed

sect

iona

lgi

rls

�21

1w

ith f

avor

able

lipi

d pr

ofile

; for

mos

t of

the

vari

able

s, f

atne

ss w

as s

light

ly g

reat

er th

an th

e in

flue

nce

of C

RF

WRN97114.qxd 3/11/06 4:37 PM Page 122

Proo

f

�� 123

Rui

z et

al.

[43]

cros

s-bo

ys�

429

9–10

yea

rsbo

ys a

nd g

irls

sect

iona

lgi

rls

�44

4C

RF

was

inve

rsel

y as

soci

ated

w

ith in

sulin

res

ista

nce,

and

sk

info

ld th

ickn

esse

s (p

�0.

001)

.C

RF

was

inve

rsel

y as

soci

ated

w

ith m

etab

olic

syn

drom

e Z

sco

re (

p�

0.02

)C

RF

was

neg

ativ

ely

asso

ciat

ed

with

TG

in g

irls

(p

�0.

026)

And

erse

n pr

ospe

ctiv

ebo

ys�

133

16–1

9 ye

ars

boys

and

gir

lset

al.

[44]

girl

s�

172

to 2

4–27

C

RF

was

ass

ocia

ted

with

car

diov

ascu

lar

year

sdi

seas

e ri

sk f

acto

rs; t

he p

roba

bilit

y fo

r ‘a

cas

e’at

the

firs

t exa

min

atio

n to

be

‘a c

ase’

at th

e se

cond

was

6.0

Bor

eham

et a

l. [4

5]pr

ospe

ctiv

ebo

ys�

251

12–1

5 –

20–2

5 bo

ys a

nd g

irls

girl

s�

203

year

sC

RF

was

inve

rsel

y as

soci

ated

with

art

eria

l st

iffn

ess

Eis

enm

ann

pros

pect

ive

boys

�36

15.9

–27

.2bo

ys a

nd g

irls

et a

l. [4

6]gi

rls

�12

year

sad

oles

cent

CR

F is

rel

ated

onl

y to

adu

lt B

MI,

W

C a

nd %

BF

(p�

0.05

).

Ferr

eira

et a

l. [4

7]pr

ospe

ctiv

ebo

ys�

175

13–3

6 ye

ars

boys

and

gir

lsgi

rls

�18

9C

RF

chan

ges

wer

e in

vers

ely

asso

ciat

ed w

ith

prev

alen

ce o

f m

etab

olic

syn

drom

e

apo

�A

polip

opro

tein

; %

BF

�pe

rcen

tage

of

body

fat

; B

MI

�bo

dy m

ass

inde

x; B

P�

bloo

d pr

essu

re;

CR

F�

card

iore

spir

ator

y fi

tnes

s;H

DL

c�

high

-den

sity

lip

opro

tein

cho

lest

erol

; L

DL

c�

low

-den

sity

lip

opro

tein

cho

lest

erol

; L

p(a)

�lip

opro

tein

(a)

; T

C�

tota

l ch

oles

tero

l;T

G�

trig

lyce

ride

s; W

C�

wai

st c

ircu

mfe

renc

e; W

/H�

wai

st to

hip

rat

io.

WRN97114.qxd 3/11/06 4:37 PM Page 123

Proo

f


fitness is negatively associated with homocysteine levels in female adolescentsafter controlling for age, puberty, birth weight, smoking, socioeconomic status,sum of six skinfolds and methylenetetrahydrofolate reductase 677C�T geno-type [48]. These findings support a previous study examining the relationshipbetween homocysteine and cardiorespiratory fitness in adults [49]. Kuo et al.

Cardiorespiratory fitness (quartiles)

4(high)

321(low)

75

80

85

90

95

100

105

110

LDLc

(mg/

dl)

MalesFemales

Fig. 5. Associations between low-density lipoprotein cholesterol (LDLc) levels andcardiorespiratory fitness quartiles in male and female Spanish adolescents. Data shown asmean and SEM [13].


4(high)

321(low)

50

55

60

65

70

75

80

85

90Tr

igly

cerid

es (m

g/d

l)

MalesFemales

*p�0.004

Fig. 4. Associations between triglycerides levels and cardiorespiratory fitness quar-tiles in male and female Spanish adolescents. Data shown as mean and SEM. * p for trend formales [13].

WRN97114.qxd 3/11/06 4:37 PM Page 124

Proo

f

�� 125

[49] showed that high homocysteine levels were negatively associated with esti-mated cardiorespiratory fitness in adult women. Moreover, one longitudinalstudy followed 499 independent community-dwelling elderly for 3 years andfound that people with elevated homocysteine levels were at an increased riskof decline in physical function [50]. However, cardiorespiratory fitness data


4(high)

321(low)

45

50

55

60

65

70

HD

Lc (m

g/d

l)

MalesFemales*p�0.013 ^p�0.045

Fig. 6. Associations between high-density lipoprotein cholesterol (HDLc) levels andcardiorespiratory fitness quartiles in male and female Spanish adolescents aged. Data shownas mean and SEM. * p for trend for males; ^ p for trend for females [13].


4(high)

321(low)

45

140

135

130

125

120

115

110

10

Apo

A-1

(mg/

dl)

MalesFemales

*p�0.028

Fig. 7. Associations between apolipoprotein (apo) A-1 levels and cardiorespiratory fit-ness quartiles in male and female Spanish adolescents aged 13–18.5 years. Data shown asmean and SEM. * p for trend for males [13].

WRN97114.qxd 3/11/06 4:37 PM Page 125

Proo

f


were not provided and a gender comparison was not performed. These resultsshould stimulate a debate on whether the metabolism of homocysteine could beone way in which the benefits of high cardiorespiratory fitness are exerted.

Cardiorespiratory fitness has also been shown to be associated with C-reactive protein and C3 in Spanish adolescents [51]. Similarly, Halle et al. [52]showed that cardiorespiratory fitness was negatively associated with low-gradeinflammation in normal weight and overweight children aged 12 years. Theyreported that interleukin 6 levels were as low for obese and fit as for lean andunfit children, while the higher interleukin 6 levels were found in the obese andunfit group. In contrast, they also showed that tumor-necrosis factor-� seemedto be primarily dependent on cardiorespiratory fitness but not obesity sincesimilar levels were found for non-obese as well as for obese children with lowcardiorespiratory fitness.

Despite the evidence on the association between cardiorespiratory fitnessand emerging and traditional cardiovascular risk factors in young and adult

1(low) 2

34

(high)

�1

�0.8

�0.6

�0.4

�0.2

0

0.2

0.4

0.6

0.8

1M

etab

olic

pro

file

Unf

avor

able

Favo

rab

le


*p�0.05

*p �0.05

OverweightNon-overweight

Fig. 8. Association between metabolic profile (computed with age- and gender-spe-cific standardized values of triglycerides, low density lipoprotein cholesterol, high densitylipoprotein cholesterol and fasting glycemia) and cardiorespiratory fitness quartiles in non-overweight and overweight Spanish adolescents. The higher is the metabolic profile thehealthier. Weight categories were constructed following the International Obesity TaskForce-proposed gender- and age-adjusted BMI cutoff points. Data shown as mean and SEM.*p for trend in both overweight and non-overweight categories [13].

WRN97114.qxd 3/11/06 4:37 PM Page 126

Proo

f

�� 127

0 0.5 1.0 1.5 2.0 2.5 3.0

Odds ratio

Low CRF

High CRF

Males

Females

Males

Females

Adjusted for age

Adjusted for age and waist circumferece

p�0.004

p�0.05

p�0.06

p�0.05

Fig. 9. Odds ratio for having an unfavorable lipid profile (triglycerides, high-densitylipoprotein cholesterol, apolipoprotein A-1, apolipoprotein B-100, total cholesterol and high-density lipoprotein cholesterol ratio) in male and female Spanish adolescents.

populations, it is still uncertain whether health criterion values for cardiorespi-ratory fitness can be identified and the implications of these from the publichealth perspective. In this respect, several health-related threshold values ofcardiorespiratory fitness have been suggested by world-wide recognized orga-nizations [53, 54]. Based on expert judgment, the European Group of PediatricWork Physiology considered a VO2max of �35 ml/kg/min for girls and�40 ml/kg/min for boys as a ‘Health Indicator’ [53]. The Cooper Institute forAerobics Research suggested �38 and �42 ml/kg/min for girls and boysrespectively as a criterion standard for the ‘Healthy Fitness Zone’ [54]. The cut-off points proposed by the Cooper Institute for adolescents were extrapolatedfrom the adults established thresholds.

Muscle Strength and Cardiovascular Risk FactorsMuscle strength refers to a balanced, healthy functioning of the muscu-

loskeletal system and requires that a specific muscle or muscle group be able togenerate force or torque. Muscle strength can also be a surrogate measure ofboth muscular endurance (that is the capacity to resist repeated contractions

WRN97114.qxd 3/11/06 4:37 PM Page 127

Proo

f


over time or maintain a maximal voluntary contraction for a prolonged periodof time), and explosive strength (that is the capacity to carry out a maximal,dynamic contraction of a muscle or muscle group).

The importance of resistance exercise in promoting health and preventingdisease has become increasingly recognized. Resistance exercise improvesskeletal muscle strength and power, but also contributes to the prevention andmanagement of atherosclerotic coronary heart disease, hypertension, diabetes,and overweight and obesity in adults. Muscle strength has been suggested to beinversely associated with all-cause mortality in men and women, independentof cardiorespiratory fitness levels [55]. However, little is known whether thehealth benefits of resistance exercise are independent of, or additive to, thosealready established for large muscle dynamic aerobic activity. Results from theAVENA study revealed significant associations between muscle strength andlow-grade inflammation. It is known that low-grade inflammation seems to playa role in the pathogenesis of atherosclerosis from early ages, suggesting thatpreventive measures should start early in life. Figure 10 shows the associationsbetween muscle strength and a compound index of low-grade inflammationintegrated by C-reactive protein and C3, according to weight categories.Regression analysis was performed on muscle strength and the logarithmic ofthis index as continuous variables separately for non-overweight and over-weight; however, in figure 10 they are broken into tertiles to illustrate the natureof the association. C-reactive protein has been recognized as cardiovascular riskfactors, and nowadays there is increasing evidence about the link between C3and cardiovascular disease.

0.2

0.4

0.6

0.8

1.0

Tertiles of muscle strenght

Low Middle High

Overweight

Normal-weight

Low

-gra

de

infla

mm

atio

n *p�0.05

Fig. 10. Associations between tertiles of muscle strength and low-grade inflammation(estimated as a compound index of C-reactive protein and C3). These results are presentedaccording to weight categories in Spanish adolescents. Weight categories were constructedfollowing the International Obesity Task Force-proposed gender- and age-adjusted bodymass index cut-off points. *p value from the regression analyses for the overweight category.

WRN97114.qxd 3/11/06 4:37 PM Page 128

Proo

f

�� 129

Taken together, these findings support the concept that cardiorespiratoryfitness and muscle strength may exert a protective effect on the cardiovascularsystem from an early age [56]. In fact, it is biologically plausible that high fitnesslevels provides more health protection than low fitness, even in healthy adoles-cents as has been found in adults. Prospective studies are needed to examine theindependent and joint effects of cardiorespiratory fitness and muscle strength inpreventing the development of cardiovascular risk factors among young peopleand adults. For public health strategies and preventive purposes it is of interestto understand the associations between diet, cardiorespiratory fitness, musclefitness and cardiovascular risk factors from early ages on.

Body Composition and Cardiovascular Risk Factors in Mediterranean Adolescence

Childhood overweight or obesity is associated with a variety of adverseconsequences both at that early age and later in life. Since childhood obesity isnow recognized as a worldwide epidemic [16] it seems relevant to study, inchildren and adolescents, the association between total body fatness and physi-cal activity and physical fitness, particularly in regions which have been tradi-tionally protected given their favorable diet and life-style. It is known that theamount of fatness is associated with a poor health status, but it is also importanthow the fat depots are distributed in the body. In fact, central body fatness isassociated with coronary heart disease morbidity and mortality and coronaryheart disease risk factors including dyslipidemia, insulin resistance and hyper-tension [57]. Most disturbances related to abdominal obesity have been estab-lished to show their onset during childhood [58]. Therefore, in this section bothtotal and central/abdominal adiposity and their relationships with physicalactivity and cardiorespiratory fitness in children and adolescents are presented.

Physical Activity, Fitness and Total Body FatTotal Body Fat in Young PopulationsDefining obesity or overweight for children and adolescents is difficult,

and there is no generally accepted definition of overweight or obesity foryouths. However, body mass index (BMI) is a widely used tool to identifyoverweight and obese children and adolescents [59]. Indeed, we have observedelevated overweight and obesity prevalence in Spanish adolescents [60], simi-lar to those observed in other European countries (including Mediterraneandiet’s countries) (fig. 3). Factors, such as socioeconomic status, seem to beinversely related to the overweight obesity prevalence. Of note is that therate of change in overweight prevalence in Spanish adolescents seems to be

WRN97114.qxd 3/11/06 4:37 PM Page 129

Proo

f


increasing [60]. Particularly in Mediterranean countries, there is an urge toestablish preventive measures to fight against this alarming increasing in thechildhood obesity epidemic. Measures to improve fitness could play a key rolenot only in obesity prevention but also in improving the health of the adults oftomorrow.

Although the BMI criterion is the most frequently used, an importantnumber of adolescents classified as overweight or obese do not have really highadiposity (32.1% of females and 42% of males) [61]. Therefore, whenever pos-sible, the anthropometric assessment of body composition should include abody fat estimation (i.e. from skinfold thickness). In this context, body fat ref-erence data from Spanish adolescents have been recently reported, helping us toclassify adolescents in comparison with a well-established reference popula-tion, and to estimate the proportion of adolescents with high or low adiposityamounts [62].

Associations of Total Body Fat with Physical Activity and FitnessA sedentary lifestyle and a significant reduction in daily physical activity are

one of the key factors of the obesity epidemic in the children and adolescents. Bycontrast, high cardiorespiratory fitness during childhood and adolescence hasbeen associated not only with a healthier cardiovascular profile during these yearsbut also later in life (table 2). For preventive purposes, it is of interest to under-stand the relative importance of the amount and intensity of physical activity notonly on total body fat but also in cardiorespiratory fitness levels. New data haveshown positive associations between physical activity, especially vigorous physi-cal activity (�6 METs) and cardiorespiratory fitness (fig. 11) [63], as well asnegative associations between vigorous physical activity and fatness in childrenand adolescents (fig. 12) [63, 64]. These results suggest that a certain level ofphysical activity needs to be achieved in order to improve the fitness and fatnessstatus. Vigorous physical activity seems to be more relevant in increasing fitnessand reducing body fat in young people. From a public health perspective thesefindings are particularly relevant.

Physical Activity, Fitness and Body Fat DistributionBody Fat Distribution in Young PopulationsThe study of fat distribution among children and adolescents is complex

because there are marked changes in circumferences and skinfold thicknessduring growth and development [65]. The two types of fat depots are abdominaland truncal fat. In population studies, the best anthropometric marker ofabdominal obesity is waist circumference. Waist circumference correlates wellwith intra-abdominal and subcutaneous fat measured by magnetic resonance

WRN97114.qxd 3/11/06 4:37 PM Page 130

Proo

f

�� 131

imaging in children and adolescents [66]. Waist circumference is a good toolfor the screening of total body fat and the metabolic syndrome. That is whywaist circumference is also a central feature of the metabolic syndrome andseveral diagnostic criteria of the condition include this marker in the definition

2.5�10 18–18 18–26 26–40 �40

2.6

2.7

2.8

2.9

3.0

3.1

3.2

Car

dio

resp

irato

ry fi

tnes

s (W

/kg)

Time spent at vigorous physical activity (min/day)

**

Fig. 11. Mean cardiorespiratory fitness stratified by time spent at vigorous physicalactivity in Swedish and Estonian children. Error bars represent 95% CIs. * A significant dif-ference was observed between those who accumulated �40 min/day of vigorous physicalactivity and those who accumulated �18 min/day at this intensity level. ^ A significant differ-ence was also observed between children who accumulated 26–40 min/day of vigorous physi-cal activity compared to those who accumulated 10–18 min/day at this level of intensity [63].

28�10 10–18 18–26 26–40 �40

44

42

40

38

36

34

32

30Sum

of f

ive

skin

fold

s (m

m)

Time spent at vigorous physical activity (min/day)

*

Fig. 12. Mean sum of five skinfolds (body fat) stratified by time spent at vigorousphysical activity in Swedish and Estonian children. Errors bars represent 95% CIs. * A sig-nificant difference was observed between those who accumulated �40 min/day of vigorousphysical activity and those who accumulated 10–18 min/day at this intensity level [63].

WRN97114.qxd 3/11/06 4:37 PM Page 131

Proo

f


[67]. In the absence of a recognized definition of increased central adiposity inyoung people, the terms ‘overweight’ and ‘obesity’ referred to central adiposityare currently being arbitrarily defined. Therefore, they have been recentlyreported reference data for waist circumference and other fat patterning indicesfrom a large sample of Spanish adolescents [Moreno et al., unpubl. data]. Thesedata, together with data from other countries, will help to establish internationalcentral obesity criteria for adolescents, giving the possibility to estimate theproportion of adolescents with high or low regional adiposity.

Associations of Body Fat Distribution with Physical Activity and FitnessIt has been reported that, even within a given BMI category, children and

adolescents with a large waist circumference are more likely to have abnormalcardiovascular disease risk factors compared to those with a small waist cir-cumference [68]. Consequently, waist circumference could be a useful tool forstudying obesity in adolescents. In adults, several studies have reported thatindividuals with better cardiorespiratory fitness have less abdominal fat and/orsmaller waist circumferences for a given BMI [69]. However, the resultsobtained so far on the relationship between physical activity and central obesityin children and adolescents are also inconsistent.

Recent results from Spanish adolescents [70] suggest that moderate tohigh levels of cardiorespiratory fitness, but not self-reported physical activity,are associated with lower abdominal adiposity, as measured by waist circum-ference (fig. 13). However, given that the questionnaire used in that study doesnot provide either the intensity level of physical activity or the frequency of

85

80

75

70

65Boys Girls

p�0.001

p�0.001

p�0.001 p for trend �0.001W

aist

circ

umfe

renc

e (c

m)

Very low CRF Low CRF High CRF Very high CRF

Fig. 13. Waist circumference (means standard error of the mean) according to car-diorespiratory fitness (CRF) quartiles in Spanish adolescents [69].

WRN97114.qxd 3/11/06 4:37 PM Page 132

Proo

f

�� 133

physical activity, it is necessary to be cautious with the physical activity-relatedconclusions from that study. Research with objective methods for measuringphysical activity, such as accelerometry, will provide accurate informationabout physical activity patterns (intensity, frequency and duration), helping toclarify this issue. In this context, data obtained from the Swedish part of theEuropean Youth Heart Study using physical activity objectively measured, hasrecently obtained the same conclusion [Ortega et al., unpubl data]. Both inchildren and adolescents, physical activity (either total, moderate or vigorous)is not associated with abdominal adiposity, as measured by waist circumfer-ence (fig. 14). This is not the case with cardiorespiratory fitness. These resultssuggest that the beneficial effects of physical activity on abdominal adipositymay be explained by its association with cardiorespiratory fitness in childrenand adolescents.

80

Wai

st c

ircum

fere

nce

(cm

)

60

40

20

Low PA Middle PA High PA

Chi

ldre

n

Gra

de

Ad

oles

cent

s

BoysGirls

Sex

Total physical activity (tertiles)

0Low PA Middle PA High PA

Total physical activity (tertiles)

80

Wai

st c

ircum

fere

nce

(cm

)

60

40

20

0

Fig. 14. Waist circumference (means) according to total physical activity (PA) inSwedish children and adolescents. Data were adjusted for age group and height. Total PA wasnot associated with waist circumference. No relationship was found between the PA intensi-ties levels (moderate, vigorous, or moderate plus vigorous) and waist circumference.

WRN97114.qxd 3/11/06 4:37 PM Page 133

Proo

f


Conclusion

Growing evidence demonstrates that the Mediterranean life-style is bene-ficial to health, especially for coronary heart disease, stroke and some forms ofcancer. Diet is one outstanding component of the Mediterranean life-style. Butit is not only diet – physical activity is another critical component. Diet andphysical activity interact in the development (or prevention) of coronary heartdisease and several other health conditions. These interactions may start earlyin life. In this way, adolescence is a critical period because it is at that timewhen the individual takes control of his/her own life-style and diet. Large life-style and dietary variations occur in different regions and countries. In many ofthem, a progressive departure from the traditional Mediterranean life-style anddiet is being observed. In Spain, a high prevalence of an unfavorable plasmalipid profile has been observed both in boys and girls. Similarly, the highprevalence of obesity in Mediterranean children and adolescents is wellknown. But it is not only physical activity. Physical fitness (especially car-diorespiratory fitness and muscle strength) is strongly associated with cardio-vascular risk factors. For public health strategies and preventive purposes it isof interest to understand the associations of diet, physical activity and fitnesson cardiovascular risk factors from early ages on. It is important to implementmeasures for preventing the deleterious consequences that these conditions aregoing to have in the health of tomorrow’s adults [71]. One positive measure isthe return to the traditional Mediterranean diet; the other is to increase the lev-els of physical activity in order to improve physical fitness. Measures toimprove fitness could play a key role in improving the health of the adults oftomorrow.

References

1 Kromhout D: Epidemiology of cardiovascular diseases in Europe. Publ Health Nutr2001;4:441–457.

2 Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJ: Global and regional burden of diseaseand risk factors, 2001: systematic analysis of population health data. Lancet 2006;367:1747–1757.

3 Kuulasmaa K, Tunstall-Pedoe H, Dobson A, Fortmann S, Sans S, Tolonen H, Evans A, Ferrario M,Tuomilehto J: Estimation of contribution of changes in classic risk factors to trends in coronary-event rates across the WHO MONICA Project populations. Lancet 2000;355:675–687.

4 Simopoulos AP, Visioli F: Mediterranean Diets. World Rev Nutr Diet. Basel, Karger, 2000, vol 87,pp 1–184.

5 Kok FJ, Kromhout D: Atherosclerosis. Epidemiological studies on the health effects of aMediterranean diet. Eur J Nutr 2004;43:I/2–I/5.

6 Trichopoulou A: Traditional Mediterranean diet and longevity in the elderly: a review. Publ HealthNutr 2004;7:943–947.

7 Tur JA, Romaguera D, Pons A: Food consumption patterns in a Mediterranean region: does theMediterranean diet still exist? Ann Nutr Metab 2004;48:193–201.

WRN97114.qxd 3/11/06 4:37 PM Page 134

Proo

f

�� 135

8 Mokdad AH, Marks JS, Stroup DF, Gerberding JL: Actual causes of death in the United States,2000. JAMA 2004;291:1238–1245.

9 Moreiras-Varela O: The Mediterranean diet in Spain. Eur J Clin Nutr 1989;43:83–87.10 Castillo-Garzon MJ: Changing the Spanish Mediterranean diet: effects on plasma lipids; in

Watson R (ed): Wild-Type Diet. New York, Humana Press, in press. 11 Delgado M, Gonzalez-Gross M, Cano MD, Gutierrez A, Castillo M: Physical exercise reverses

diet-induced increases in LDL-cholesterol and apo B levels in healthy ovo-lactovegetarian sub-jects. Nutr Res 2000;20:1707–1714.

12 Ruiz JR, Mesa JL, Mingorance I, Rodriguez-Cuartero A, Castillo MJ: Sports requiring stress-ful physical exertion cause abnormalities in plasma lipid profile. Rev Esp Cardiol 2004;57: 499–506.

13 Mesa JL, Ruiz J, Ortega FB, Warnberg J, Gonzalez-Lamuno D, Moreno LA, Gutierrez A, CastilloMJ: Aerobic physical fitness in relation to blood lipids and fasting glycaemia in adolescents: influ-ence of weight status. Nutr Metab Cardiovasc Dis 2006;285–293.

14 Gonzalez-Gross M, Castillo MJ, Moreno L, Moreno L, Nova E, Gonzalez-Lamuno D, Perez-Llamas F, Gutierrez A, Garaulet M, Joyanes M, Leiva A, Marcos A: Feeding and assessment ofnutritional status of spanish adolescents (AVENA study). Evaluation of risks and interventionalproposal. I. Methodology. Nutr Hosp 2003;18:15–28.

15 Ruiz JR, Ortega FB, Moreno LA, Wärnberg J, Gonzalez-Gross M, Cano MD, Gutierrez A, Castillo MJ, the AVENA Study Group: Serum lipid and lipoprotein reference values of Spanishadolescents; The AVENA study. Soz Praventiv Med 2006;51:99–109.

16 Flodmark CE, Lissau I, Moreno LA, Pietrobelli A, Widhalm K: New insights into the field of chil-dren and adolescents’ obesity: the European perspective. Int J Obes 2004;28:1189–1196.

17 Castillo MJ, Ruiz JR, Ortega FB, Gutierrez A: Anti-aging therapy through fitness enhancement.Clin Intervent Aging 2006;1:213–220.

18 Carnethon MR, Gidding SS, Nehgme R, Sidney S, Jacobs DR Jr, Liu K: Cardiorespiratory fitnessin young adulthood and the development of cardiovascular disease risk factors. JAMA2003;290:3092–3100.

19 Laukkanen JA, Lakka TA, Rauramaa R, Kuhanen R, Venalainen JM, Salonen R, Salonen JT: Cardiovascular fitness as a predictor of mortality in men. Arch Intern Med 2001;161:825–831.

20 Balady GJ: Survival of the fittest: more evidence. N Engl J Med 2002;346:852–854.21 Kurl S, Laukkanen JA, Rauramaa R, Lakka TA, Sivenius J, Salonen JT: Cardiorespiratory fitness

and the risk for stroke in men. Arch Intern Med 2003;163:1682–1688.22 Gulati M, Pandey DK, Arnsdorf MF, Lauderdale DS, Thisted RA, Wicklund RH, Al-Hani AJ,

Black HR: Exercise capacity and the risk of death in women: the St James Women Take HeartProject. Circulation 2003;108:1554-–1559.

23 Mora S, Redberg RF, Cui Y, Whiteman MK, Flaws JA, Sharrett AR, Blumenthal RS: Ability ofexercise testing to predict cardiovascular and all-cause death in asymptomatic women: a 20-yearfollow-up of the lipid research clinics prevalence study. JAMA 2003;290:1600–1607.

24 Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood JE: Exercise capacity and mortal-ity among men referred for exercise testing. N Engl J Med 2002;346:793–801.

25 Lee CD, Blair SN: Cardiorespiratory fitness and smoking-related and total cancer mortality inmen. Med Sci Sports Exerc 2002;34:735–739.

26 Evenson KR, Stevens J, Cai J, Thomas R, Thomas O: The effect of cardiorespiratory fitness andobesity on cancer mortality in women and men. Med Sci Sports Exerc 2003;35:270–277.

27 Lee CD, Folsom AR, Blair SN: Physical activity and stroke risk. Stroke 2003;34:2475–2481.28 Sawada SS, Muto T, Tanaka H, Lee IM, Paffenbarger RS Jr, Shindo M, Blair SN: Cardio-

respiratory fitness and cancer mortality in Japanese men: a prospective study. Med Sci SportsExerc 2003;35:1546–1550.

29 Seibaek M, Vestergaard H, Burchardt H, Sloth C, Torp-Pedersen C, Nielsen SL, Hildebrandt P,Pedersen O: Insulin resistance and maximal oxygen uptake. Clin Cardiol 2003;26:515–520.

30 Bertoli A, Di Daniele N, Ceccobelli M, Ficara A, Girasoli C, De Lorenzo A: Lipid profile, BMI,body fat distribution, and aerobic fitness in men with metabolic syndrome. Acta Diabetol 2003;40:130S–133S.

WRN97114.qxd 3/11/06 4:37 PM Page 135

Proo

f


31 Lakka TA, Laaksonen DE, Lakka HM, Mannikko N, Niskanen LK, Rauramaa R, Salonen JT:Sedentary lifestyle, poor cardiorespiratory fitness, and the metabolic syndrome. Med Sci SportsExerc 2003;35:1279–1286.

32 Colcombe SJ, Erickson KI, Raz N, Webb AG, Cohen NJ, McAuley E, Kramer AF: Aerobic fitnessreduces brain tissue loss in ageing humans. J Gerontol [A] 2003;58:176–180.

33 Barnes DE, Yaffe K, Satariano WA, Tager IB: A longitudinal study of cardiorespiratory fitness andcognitive function in healthy older adults. J Am Geriatr Soc 2003;51:459–465.

34 Metter EJ, Talbot LA, Schrager M, Conwit R: Skeletal muscle strength as a predictor of all-causemortality in healthy men. J Gerontol [A] 2002;57:B359–B365.

35 Seguin R, Nelson ME: The benefits of strength training for older adults. Am J Prev Med2003;25:S141–S149.

36 Ruiz JR, España-Romero V, Ortega FB, Sjostrom M, Castillo MJ, Gutierrez A: Hand span influ-ences optimal grip span in male and female teenagers. J Hand Surg Am. 2006;31:1367–1372.

37 Ruiz JR, Mesa JL, Castillo MJ, Gutierrez A: Hand size influences optimal grip span in women butnot in men. J Hand Surg 2002;27:897–901.

38 Gutin B, Yin Z, Humphries MC, Hoffman WH, Gower B, Barbeau P: Relations of fatness and fit-ness to fasting insulin in black and white adolescents. J Pediatr 2004;145:737–743.

39 Brage S, Wedderkopp N, Ekelund U, Franks PW, Wareham NJ, Andersen LB, Froberg K:European Youth Heart Study (EYHS): Features of the metabolic syndrome are associated withobjectively measured physical activity and fitness in Danish children: the European Youth HeartStudy (EYHS). Diabetes Care 2004;27:2141–2148.

40 Reed KE, Warburton DE, Lewanczuk RZ, Haykowsky MJ, Scott JM, Whitney CL, McGavock JM,McKay HA: Arterial compliance in young children: the role of aerobic fitness. Eur J CardiovascPrev Rehabil 2005;12:492–497.

41 Eisenmann JC, Katzmarzyk PT, Perusse L, Tremblay A, Despres JP, Bouchard C: Aerobic fitness,body mass index, and CVD risk factors among adolescents: the Quebec family study. Int J ObesRelat Metab Disord 2005;29:1077–1083.

42 Gutin B, Yin Z, Humphries MC, Bassali R, Le NA, Daniels S, Barbeau P: Relations of body fat-ness and cardiovascular fitness to lipid profile in black and white adolescents. Pediatr Res2005;58:78–82.

43 Ruiz JR, Ortega FB, Meusel D, Harro M, Oja P, Sjöström M: Cardiorespiratory fitness is associ-ated with features of metabolic risk factors in children. Should cardiorespiratory fitness beassessed in a European health monitoring system? The European Youth Heart Study. J Publ Health2006;14:94–102.

44 Andersen LB, Hasselstrøm H, Grønfeldt V, Hansen SE, Froberg K: The relationship between phys-ical fitness and clustered risk, and tracking of clustered risk from adolescence to young adulthood:eight years follow-up in the Danish Youth and Sport Study. Int J Behav Nutr Phys Fitness 2004;1:6.

45 Boreham CA, Ferreira I, Twisk JW, Gallagher AM, Savage MJ, Murray LJ: Cardiorespiratory fit-ness, physical activity, and arterial stiffness: The Northern Ireland Young Hearts Project.Hypertension 2004;44:721–726.

46 Eisenmann JC, Wickel EE, Welk GJ, Blair SN: Relationship between adolescent fitness and fat-ness and cardiovascular disease risk factors in adulthood: the Aerobics Center Longitudinal Study(ACLS). Am Heart J 2005;149:46–53.

47 Ferreira I, Henry RM, Twisk JW, van Mechelen W, Kemper HC, Stehouwer CD: The metabolicsyndrome, cardiopulmonary fitness, and subcutaneous trunk fat as independent determinants ofarterial stiffness: the Amsterdam Growth and Health Longitudinal Study. Arch Intern Med2005;165:875–882.

48 Ruiz JR, Sola R, Gonzalez-Gross M, Ortega FB, Vicente-Rodriguez G, Garcia-Fuentes M,Gutierrez A, Sjöström M, Pietrzik K, Castillo MJ: Cardiovascular fitness is negatively associatedwith homocysteine levels in female adolescents. Arch Pediatr Adolesc Med. In press.

49 Kuo HK, Yen CJ, Bean JF: Levels of homocysteine are inversely associated with cardiovascularfitness in women, but not in men: data from the National Health and Nutrition ExaminationSurvey 1999–2002. J Intern Med 2005;258:328–335.

50 Kado DM, Bucur A, Selhub J, Rowe JW, Seeman T: Homocysteine levels and decline in physicalfunction. Studies of successful aging. Am J Med 2002;113:537–542.

WRN97114.qxd 3/11/06 4:37 PM Page 136

Proo

f

�� 137

51 Wärnberg J: Inflammatory Status in Adolescents; the Impact of Health Determinants Such AsOverweight and Fitness. Thesis, Karolinska University Press, Stockholm, 2006.

52 Halle M, Berg A, Northoff H, Keul J: Importance of TNF-alpha and leptin in obesity andinsulin resistance: a hypothesis on the impact of physical exercise. Exerc Immunol Rev 1998;4:77–94.

53 Bell RD, Macek M, Rutenfranz J, Saris WHM: Health indicators and risk factors of cardiovasculardiseases during childhood and adolescence; in Rutenfranz J, Mocelin R, Klimt F (eds): Childrenand Exercise. Part XII. Champaign, Human Kinetics, 1986, pp 19–27.

54 Cooper Institute for Aerobics Research: FITNESSGRAM Test Administration Manual.Champaign, Human Kinetics, 1999.

55 Jurca R, Lamonte MJ, Barlow CE, Kampert JB, Church TS, Blair SN: Association of muscularstrength with incidence of metabolic syndrome in men. Med Sci Sports Exerc 2005;37:1849–1855.

56 Ruiz JR, Ortega FB, Gutierrez A, Meusel D, Sjöström M, Castillo MJ, on behalf of the HELENAStudy Group: Health-related fitness assessment in childhood and adolescence: a Europeanapproach based on the AVENA, EYHS and HELENA studies. J Publ Health 2006;14:269–277.

57 Rexrode KM, Carey VJ, Hennekens CH, Walters EE, Colditz GA, Stampfer MJ, Willett WC,Manson JE: Abdominal adiposity and coronary heart disease in women. JAMA 1998;280:1843–1848.

58 Freedman DS, Serdula MK, Srinivasan SR, Berenson GS: Relation of circumferences and skin-fold thicknesses to lipid and insulin concentrations in children and adolescents: the BogalusaHeart Study. Am J Clin Nutr 1999;69:308–317.

59 Cole TJ, Bellizzi MC, Flegal KM, Dietz WH: Establishing a standard definition for child over-weight and obesity worldwide: international survey. BMJ 2000;320:1240–1243.

60 Moreno LA, Mesana MI, Fleta J, Ruiz JR, Gonzalez-Gross M, Sarria A, Marcos A, Bueno M,Group AS: Overweight, obesity and body fat composition in spanish adolescents. The AVENAStudy. Ann Nutr Metab 2005;49:71–76.

61 Rodriguez G, Moreno LA, Blay MG, Blay VA, Garagorri JM, Sarria A, Bueno M: Body composi-tion in adolescents: measurements and metabolic aspects. Int J Obes 2004;3:S54–S58.

62 Moreno LA, Mesana MI, Gonzalez-Gross M, Gil CM, Fleta J, Warnberg J, Ruiz JR, Sarria A,Marcos A, Bueno M: Anthropometric body fat composition reference values in Spanish adoles-cents. The AVENA Study. Eur J Clin Nutr 2006;60:191–196.

63 Ruiz JR, Rizzo NS, Hurtig-Wennlöf A, Ortega FB, Warnberg J, Sjöström M: Relations of totalphysical activity and intensity to fitness and fatness in children: The European Youth Heart Study.Am J Clin Nutr 2006;84:298–302.

64 Gutin B, Yin Z, Humphries MC, Barbeau P: Relations of moderate and vigorous physical activityto fitness and fatness in adolescents. Am J Clin Nutr 2005;81:746–750.

65 Ruiz JR, Ortega FB, Moreno LA, Wärnberg J, Mesa JL, Gonzalez-Gross M, Tresaco B, Cano MD,Marcos A, Gutierrez A, Castillo MJ, the AVENA Study Group: Serum lipid and lipoprotein profileduring pubertal development in Spanish adolescents aged 13–18.5 years. The AVENA Study.Horm Metabol Res. In press.

66 Brambilla P, Bedogni G, Moreno LA, Goran MI, Gutin B, Fox KR, Peters DM, Barbeau P, DeSimone M, Pietrobelli A: Crossvalidation of anthropometry against magnetic resonance imagingfor the assessment of visceraI and subcutaneous adipose tissue in children. Int J Obes 2006;30:23–30.

67 Bueno G, Bueno O, Moreno LA, García R, Tresaco B, Garagorri JM: Diversity of metabolic syn-drome risk factors in obese children and adolescents. J Physiol Biochem. In press.

68 Janssen I, Katzmarzyk PT, Srinivasan SR, Chen W, Malina RM, Bouchard C, Berenson GS:Combined influence of body mass index and waist circumference on coronary artery disease riskfactors among children and adolescents. Pediatrics 2005;115:1623–1630.

69 Janssen I, Katzmarzyk PT, Ross R: Waist circumference and not body mass index explains obe-sity-related health risk. Am J Clin Nutr 2004;79:379–384.

70 Ortega FB, Tresaco B, Ruiz JR, Moreno LA, Martin-Matillas M, Mesa JL, Warnberg J, Bueno M,Tercedor P, Gutiérrez A, Castillo MJ: Cardiorespiratory fitness is associated with favorableabdominal adiposity in adolescents. Obes Res. In press.

WRN97114.qxd 3/11/06 4:37 PM Page 137

Proo

f


71 Castillo Garzon MJ, Ortega Porcel FB, Ruiz J: Improvement of physical fitness as anti-aging inter-vention. Med Clin 2005;124:146–155.

Prof. Manuel J. Castillo-GarzonDepartment of Physiology, School of MedicineUniversity of GranadaES–18071 Granada (Spain)Tel. 34 958 243540, Fax 34 958 249015, E-Mail [email protected]

WRN97114.qxd 3/11/06 4:37 PM Page 138

183

Conclusiones

I. Se han descrito los valores de referencia para los niveles de lípidos y lipoproteínas sanguíneas en adolescentes españoles, hallando que un número elevado de los mismos presenta un perfil lipídico poco saludable.

II. El desarrollo madurativo en el que se encuentra el adolescente influencia el perfil lipídico así como la cantidad y distribución de la grasa corporal.

III. El nivel de condición física se relaciona con parámetros de salud de niños y adolescentes.

IV. La capacidad aeróbica en niños de 9 y 10 años se asocia inversamente con factores tradicionales de riesgo cardiovascular, tales como el perfil lipídico, resistencia a la insulina y masa grasa.

V. La capacidad aeróbica en niñas adolescentes se asocia inversamente con factores noveles de riesgo cardiovascular, tales como el nivel de homocisteína, y esto tras controlar por diversos factores de confusión incluido el genotipo MTHFR 677C>T.

VI. La fuerza muscular se relaciona inversamente con parámetros de inflamación. Los patrones de estas asociaciones son más relevantes en adolescentes son sobrepeso.

VII. Se ha desarrollado y validado una fórmula para estimar la capacidad aeróbica basada en los modelos de redes neuronales construida a partir de: test de ida y vuelta de 20 metros, la edad, el sexo, la talla y el peso del adolescente.

VIII. La fuerza de prensión manual en adolescentes está influenciada por el tamaño de la mano y el tamaño del agarre de dinamómetro.

IX. Los datos publicados en la literatura científica reclaman la necesidad de desarrollar, evaluar e implementar estrategias de prevención en Salud Pública haciendo especial hincapié en la mejora de la condición física.

Conclusión general:

Los resultados de la presente memoria de Tesis ponen de manifiesto la importancia y utilidad de la valoración de la condición física como un determinante de salud que puede ser utilizado en instituciones sanitarias y educativas como una estrategia más para la prevención de enfermedades cardiovasculares en la vida adulta.

184

Conclusions

I. The reference values regarding the distribution of serum lipid and lipoprotein levels of Spanish adolescents are presented. A high number of subjects had an unhealthy lipid profile.

II. The assessment of pubertal development may provide additional valuable information when interpreting serum lipid profile and body fat in adolescents.

III. Physical fitness is a key health marker in children and adolescents.

IV. Cardiorespiratory fitness is inversely associated with traditional cardiovascular disease risk factors, such as serum lipid profile, insulin resistance and body fat, in children aged 9 to 10 years.

V. Cardiorespiratory fitness is inversely associated with plasma homocysteine levels in female adolescents after controlling for potential cofounders including MTHFR 677C>T genotype.

VI. Low-grade inflammation is negatively associated with muscle strength in adolescents. The patterns of these associations seem more relevant in overweight adolescents.

VII. An artificial neural network-based equation to estimate VO2max from 20m shuttle run test performance (last half stage completed), sex, age, weight and height in adolescents has been developed and cross-validated.

VIII. There is an optimal grip span to which the dynamometer should be adjusted when measuring handgrip strength in adolescents.

IX. Scientific data demonstrate that there is an urgent need for the development, testing and implementation of preventive strategies in Public Health with stronger emphasis on physical fitness.

Overall conclusion:

The results of the present work highlight the importance and usefulness of measuring physical fitness. Physical fitness should be measured in schools and included in the Health Monitoring Systems.

185

Curriculum Vitae abreviado [Short CV]

Actividad académica

Diplomado en Magisterio, especialidad Educación Física. Universidad de Granada,

Facultad de Ciencias de la Educación (Junio 1999).

Licenciado en Ciencias de la Actividad Física y del Deporte. Universidad de Granada,

Facultad de Ciencias de la Actividad Física y del Deporte (Junio 2002).

Estancia de estudios de Licenciatura en la John Moores University, Liverpool, Inglaterra

(Enero a Mayo de 2002).

Estudios de Doctorado, Universidad de Granada, España (Octubre 2002 a Febrero

2007).

Estancia de investigación en el Departamento de Biosciences and Nutrition, Karolinska

Institutet, Suecia (Julio 2005 a Diciembre 2005, Agosto 2006 a Diciembre 2006).

Profesor invitado en la Facultad de Medicina y Facultad de Ciencias de la Actividad

Física y del Deporte, Universidad de Granada (desde curso académico 2002/2003).

Profesor invitado en el Master of Public Health, Unit for Preventive Nutrition, Department

of Biosciences and Nutrition at NOVUM, Karolinska Institutet, Sweden (curso académico

2005/2006 y 2006/2007).

Artículos científicos

Revistas Internacionales contempladas en el JCR

1. Gonzalez-Gross M, Gutierrez A, Mesa JL, Ruiz-Ruiz J, Castillo MJ. Nutrition in the sport practice: adaptation of the food guide pyramid to the characteristics of athletes’ diet. Arch Latinoamer Nutr 2001; 51: 321-331.

2. Ruiz-Ruiz J, Mesa JL, Castillo MJ, Gutierrez A. Hand size influences optimal grip span in women but not in men. J Hand Surg [Am] 2002; 27: 897-901.

3. Mesa JL, Ruiz JR, Gonzalez-Gross M, Gutierrez A, Castillo MJ. Creatine supplementation and skeletal muscle metabolism in physical exercise. Sports Med2002; 32: 903-944.

4. Gonzalez-Gross M, Ruiz JR, Moreno LA, de Rufino-Rivas P, Garaulet M, Mesana MI, Gutierrez A and the AVENA group. Body composition and physical performance of Spanish adolescents. The AVENA pilot study. Acta Diabetol 2003; 40: S299-S301.

5. Gutierrez A, Mesa JL, Ruiz JR, Chirosa LJ, Castillo MJ. Sauna-induced rapid weight loss decreases explosive power in women but not in men. Int J Sports Med 2003; 24: 518-522.

6. Gutierrez A, Gonzalez-Gross M, Ruiz JR, Mesa JL, Castillo MJ. Exposure to hypoxia decreases growth hormone response to physical exercise in untrained subjects. JSports Med Phys Fitness 2003; 43: 554-558.

7. Castillo MJ, Ruiz JR. Use of Short Message Service (SMS) of cell phone to provide feedback in teaching and learning process. British Medical Journalhttp://bmj.com/cgi/eletters/326/7386/437#35188 31 July 2003.

8. Ruiz JR, Mesa JL, Mingorance I, Rodriguez Cuartero A, Castillo MJ. [Sports requiring stressful physical exertion cause abnormalities in plasma lipid profile]. Rev Esp Cardiol 2004; 57: 499-506.

9. Ortega Porcel F, Ruiz-Ruiz J, MJ Castillo Garzon, A Gutierrez Sainz. Hiponatremia en esfuerzos de ultraresistencia: efectos sobre la salud y el rendimiento. ArchLatinoamer Nutr 2004; 52: 155-164.

186

10. Castillo Garzon MJ, Ortega Porcel FB, Ruiz-Ruiz J. Improvement of physical fitness as anti-aging intervention. Med Clin 2005; 124: 146-155.

11. Moreno LA, MI Mesana, Fleta J, Ruiz JR, M Gonzalez-Gross, A Sarria, A Marcos, M Bueno, and the AVENA Study Group. Overweight, obesity and body fat composition in Spanish adolescents. The AVENA Study. Ann Nut Metab 2005; 49: 71-76.

12. Ortega FB, Ruiz JR, Castillo MJ, Moreno LA, Gonzalez-Gross M, Wanberg J, Gutierrez A y grupo AVENA. [Low level of physical fitness in Spanish adolescents. Relevance for future cardiovascular health (AVENA Study)]. Rev Esp Cardiol2005, 58: 898-909.

13. Ruiz JR, Ortega F, Gutierrez A, Castillo MJ, Agil A. Increased susceptibility to plasma lipid peroxidation in untrained subjects after an extreme mountain bike challenge at moderate altitude Int J Sports Med 2006; 27: 587-589.

14. Moreno LA, Mesana MI, González-Gross M, Gil CM, Fleta J, Wärnberg J, Ruiz JR,Sarria A, Marcos A, Bueno M and the AVENA Study Group. Anthropometric body fat composition reference values in Spanish adolescents. The AVENA Study. Eur J Clinical Nutr 2006; 60: 191-196.

15. Ruiz JR, Ortega FB, Moreno LA, Wärnberg J, Gonzalez-Gross M, Cano MD, Gutierrez A, Castillo MJ, and the AVENA Study Group. Serum lipid and lipoprotein reference values of Spanish adolescents; The AVENA study. Soz Praventiv Med 2006; 51: 99-109.

16. Mesa JL, Ruiz JR, Ortega FB, Warnberg J, Gonzalez-Lamuño D, Moreno LA, Gutierrez A, Castillo MJ, and the AVENA Study Group. Aerobic physical fitness in relation to blood lipids and fasting glycaemia in adolescents. Influence of weight Status. Nutr Metab Cardiovasc Dis 2006; 16: 285-293.

17. Ruiz JR, Ortega FB, Meusel D, Harro M, Oja P, Sjöström M. Cardiorespiratory fitness is associated with features of metabolic risk factors in children. Should cardiorespiratory fitness be assessed in a European health monitoring system? The European Youth Heart Study. J Public Health 2006; 14: 94-102.

18. Mesa JL, Ortega FB, Ruiz JR, Castillo MJ, Hurtig Wennlöf A, Gutierrez A. The importance of cardiorespiratory fitness for healthy metabolic traits in children and adolescents. The AVENA Study. J Public Health 2006; 14: 178-180.

19. Ruiz JR, Rizzo NS, Hurtig-Wennlöf A, Ortega FB, Warnberg J, Sjöström M. Relations of total physical activity and intensity to fitness and fatness in children; The European Youth Heart Study. Am J Clin Nutr 2006; 84: 299-303.

20. Warnberg J, Nova E, Moreno LA, Romeo J, Mesana MI, Ruiz JR, Ortega FB, Sjöström M, Bueno M, Marcos A; AVENA Study Group. Inflammatory proteins are related to total and abdominal adiposity in a healthy adolescent population: the AVENA Study. Am J Clin Nutr. 2006; 84: 505-512.

21. Ortega FB, Ruiz JR, Gutierrez A, Castillo MJ. Extreme mountain bike challenges may induce sub-clinical myocardial damage. J Sports Med Phys Fitness. 2006; 46: 489-493.

22. Castillo Garzon MJ, Ruiz JR, Ortega FB, Gutierrez A. Anti-aging therapy through fitness enhancement. Clinical Interventions in Aging 2006; 1: 213-220.

23. Ruiz JR, España-Romero V, Ortega FB, Sjöström M, Castillo MJ, Gutiérrez A. Hand span influences optimal grip span in male and female teenagers. J Hand Surgery[Am] 2006; 31: 1367-1372.

24. Ruiz JR, Ortega FB, Gutierrez A, Sjöström M, Castillo MJ. Health-related physical fitness assessment in childhood and adolescence; A European approach based on the AVENA, EYHS and HELENA studies. J Public Health 2006; 14: 269-277.

25. Grjibovski AM, Bergman P, Hagströmer M, Hurtig-Wennlöf A, Meusel D, Ortega FB, Patterson E, Poortvliet E, Rizzo N, Ruiz JR, Wärnberg J, Sjöström M. A dropout analysis of the second phase of the Swedish part of the European Youth Heart Study. J Public Health 2006; 14: 261-268.

187

26. Ruiz JR, Ortega FB, Tresaco B, Warnberg J, Mesa JL, Gonzalez-Gross M, Moreno LA, Marcos A, Gutierrez A, Castillo MJ. Serum lipids, body mass index and waist circumference during pubertal development in Spanish adolescents: The AVENA Study. Horm Metab Res 2006; 38: 832-837.

27. Mesa JL, Ortega FB, Ruiz JR, Castillo MJ, Tresaco B, Carreño F, Moreno LA, Gutierrez A, Bueno M. Anthropometric determinants of a clustering of lipid-related metabolic risk factors in overweight and non-overweight adolescents-influence of cardiorespiratory fitness. The AVENA Study. Ann Nutr Metab 2006; 50: 519-527.

28. Castillo-Garzon MJ, Ruiz JR, Ortega FB, Gutierrez-Sainz A. A Mediterranean diet is not enough for health: Physical fitness is an important additional contributor to health for the adults of tomorrow. World Rev Nutr Diet 2007; 97: 114-138.

29. Tercedor P, Martin-Matillas M, Chillon P, Perez Lopez IJ, Ortega FB, Warngberg J, Ruiz JR, Delgado M y Grupo AVENA. Incremento del consumo de tabaco y disminucion del nivel de practica de actividad fisica en adolescentes españoles. Estudio AVENA. Nutr Hosp 2007; 22: 97-102.

30. Ruiz JR, Sola R, Gonzalez-Gross M, Ortega FB, Vicente-Rodriguez G, Garcia-Fuentes M, Gutierrez A, Sjöström M, Pietrzik K, Castillo MJ. Cardiovascular fitness is negatively associated with homocysteine levels in female adolescents. ArchPediatr Adol Med 2007; 161: 166-171.

31. Ruiz JR, Hurtig-Wennlöf A, Ortega FB, Patterson E, Nilsson T, Castillo MJ, Sjöström M. Homocysteine levels in children and adolescents are associated with the methylenetetrahydrofolate reductase 677C > T genotype, but not with physical activity, fitness or fatness: The European Youth Heart Study. Brit J Nutrition inpress.

32. Ruiz JR, Ortega FB, Rizzo NS, Villa L, Hurtig-Wennlöf A, Oja L, Sjostrom M. High cardiorespiratory fitness is associated with low metabolic risk score in children; The European Youth Heart Study Pediatrics Res in press.

33. Rizzo N, Ruiz JR, Hurtig-Wennlöf A, Ortega FB, Sjöström M. Relationship of physical activity, fitness and fatness with features of metabolic syndrome in Swedish children and adolescents; The European Youth Heart Study. J Pediatrics in press.

34. Hurtig-Wennlöf A, Ruiz JR, Harro M, Sjöström M. Cardiorespiratory fitness relates more strongly than physical activity to cardiovascular disease risk factors in healthy children and adolescents. The European Youth Heart Study. Eur J Cardiovasc Prev Rehabil in press.

35. Ortega FB, Tresaco B, Ruiz JR, Moreno LA, Martín-Matillas M, Mesa JL, Warnberg J, Bueno M, Tercedor P, Gutiérrez A, Castillo MJ. Cardiorespiratory fitness is associated with favorable abdominal adiposity in adolescents. The AVENA study. Obesity in press.

36. Ortega FB, Ruiz JR, Mesa JL, Gutierrez A, Sjöström M. Cardiovascular fitness in adolescents: the influence of sexual maturation status. The AVENA and EYHS studies. Am J Hum Biol in press.

188

Revistas Nacionales e Internacionales no contempladas en el JCR

37. Mesa JL, Ruiz JR, Castillo MJ, Gutiérrez A. Hacia un logismo del entrenamiento deportivo. Lecturas EF Deportes. Revista digital 2001; 38. http://www.efdeportes.com/efd38/entren.htm

38. Mesa JLM, Ruiz JR, Hernández J, Mula FJ, Castillo MJ, Gutiérrez A. Creatina como ayuda ergogénica: efectos adversos. Archivos de Medicina del Deporte 2001; 86: 613-619.

39. Ruiz JR, Mula FJ, Mesa JL, Castillo MJ, Gutierrez A. Mejora de la fuerza por supercompensación tras sobrecarga y descanso activo. Revista de Entrenamiento Deportivo 2002; 16: 5-12.

40. Ruiz JR, Mesa JL, Castillo MJ, Gutierrez A. La fatiga de la musculatura antagonista implicada en el pedaleo mejora el rendimiento en ciclismo. Revista de Entrenamiento Deportivo 2002; 16: 5-8.

41. Ruiz JR, Mesa JLM, Mula FJ, Castillo MJ, Gutierrez A. Hidratación y rendimiento: pautas para una elusión efectiva de la deshidratación por ejercicio. Apunts,Educación Física y Deportes 2003; 70: 26-33.

42. Ruiz JR, Canovas J, Capitan LM, Imbroda J, Candel J. Spiribol®: un giro al deporte. Lecturas EF Deportes. Revista digital 2003, Año 9, 64. http://www.efdeportes.com/efd64/spiribol.htm

43. Ruiz JR, Gutierrez A, Ortega F, Castillo MJ. El ejercicio físico como terapia antienvejecimiento. Medicina Estética y Longevidad 2003; 9: 22-23.

44. Ortega F, Ruiz J, Rodriguez G, Gutierrez A, Castillo M. Physical fitness evaluation-interpretation software. Int J Comp Sci Sport 2003; 2: 160-162.

45. Albert F, Ortega F, Ruiz J, Castillo M, Gutierrez A. Software for anthropometric assessment providing indexes of interest for health and sport. Int J Comp Sci Sport 2003;2:142-144.

46. Zabala M, Ruiz JR, Mesa JLM, Gutierrez A. Adaptaciones fisiológica del entrenamiento en altitud. Experiencia del equipo nacional de mountain bike en el mundial de Colorado. Revista de Entrenamiento Deportivo 2004; 18: 13-22.

47. Ortega Porcel F, Chillon Garzon P, Ruiz-Ruiz J, Delgado Fernandez M, Moreno Aznar LA, Castillo Garzon MJ, Gutierrez Sainz A. Un Programa de Intervención Nutricional y Actividad Física de Seis Meses Produce Efectos Positivos Sobre la Composición Corporal de Adolescentes Escolarizados. Rev Esp Pediatr 2004; 60: 293-290.

48. Ruiz JR, Gonzalez-Gross M, Mesa JLM. Aspectos fisiológico-deportivos del futbolista de élite. Selección, Revista Española e Iberoamericana de Medicina de la Educación Física y el Deporte 2004; 13: 124-129.

49. Cerezo S, Ruiz JR, Ortega FB, Albert F, Sola R, Castillo MJ, Gutiérrez A. Efecto de la altitud sobre la deshidratación y el rendimiento físico tras un ejercicio prolongado en sujetos entrenados. Selección, Revista Española e Iberoamericana de Medicina de la Educación Física y el Deporte 2005; 14: 3-9.

50. Wärnberg J, Ruiz JR, Ortega FB, Romeo J, Gonzalez-Gross M, Moreno LA, Garcia-Fuentes M, Gomez S, Nova E, Diaz LE, Marcos A y grupo AVENA. Estudio AVENA* (Alimentación y valoración del estado nutricional en adolescentes). Resultados obtenidos 2003-2006. Pediatría Integral 2006; 1: S50-55.

51. Carreño Galvez F, Garcia Artero E, Ortega FB, Ruiz JR, Gutierrez A. Factores que afectan a la economía de carrera (I). Revista de Entrenamiento Deportivo 2006; 1: 13-18.

52. Carreño Galvez F, Garcia Artero E, Ortega FB, Ruiz JR, Gutierrez A. Factores que afectan a la economía de carrera (II). Revista de Entrenamiento Deportivo 2006; 2: 13-19.

189

53. Garcia-Artero E, Ortega Porcel FB, Ruiz Ruiz J, Carreño Galvez F. Entrenamiento vibratorio. Base fisiológica y efectos funcionales. Selección, Revista Española e Iberoamericana de Medicina de la Educación Física y el Deporte 2006; 15: 78-86.

54. España-Romero V, Ortega Porcel FB, Garcia Artero E, Ruiz JR, Gutierrez Sainz A. Performance, anthropometric and muscle strength characteristics in spanish elite rock climbers. Selección, Revista Española e Iberoamericana de Medicina de la Educación Física y el Deporte en prensa.

Premios recibidos

1. Premio de Libros a los 200 Mejores Rendimientos Académicos, Curso 1998/1999. Sección de Becas Propias de la Universidad de Granada. Universidad de Granada. 19 de Octubre de 1999.

2. Premio de Libros a los 200 Mejores Rendimientos Académicos, Curso 1999/2000. Sección de Becas Propias de la Universidad de Granada. Universidad de Granada. 22 de Enero de 2001.

3. Primer premio en la modalidad de comunicación oral. Federación Andaluza de Fútbol y CEDIFA. 30 de Noviembre y 1 de Diciembre de 2001. Sevilla. La aplicación de un ciclo de recuperación tras seis días de sobrecarga mejora el rendimiento futbolista. VII Jornadas Andaluzas de Salud e Investigación en el Fútbol.

4. Segundo premio en la modalidad de póster. III Congreso Internacional de Educación Física e Interculturalidad. Universidad Politécnica de Cartagena, Cartagena. 14-17 de Noviembre de 2002. En busca de un índice de condición física fiable en adolescentes: estudio piloto de AVENA.

5. 3er

Premio Nacional Fin de Carrera de Educación Universitaria. Curso Académico 2001-2002. BOE 170 de 17 de Julio 2003. Ministerio de Educación, Cultura y Deporte. ORDEN ECD/2008/2003, de 30 de mayo, por la que se adjudican los Premios Nacionales de Fin de Carrera de Educación Universitaria correspondientes al curso académico 2001/2002.

6. Mención Honorífica por el Vicerrectorado de Planificación, Calidad y Evaluación Docente de la Universidad de Granada. Utilización del sistema de mensajes cortos (SMS) para mejorar la calidad del proceso enseñanza-aprendizaje en la universidad. 30 de Junio de 2003.

7. Premio a los 10 mejores artículos sometidos por investigadores jóvenes al 9th European Nutrition Conference, Rome, 1-4 de Octubre de 2003. Simple physical assessment for lipid disturbances screening in adolescents: The AVENA study"(reference number 438).

8. Segundo accésit de los premios de la Sociedad Española de Cardiología a los mejores artículos publicados en Revista Española de Cardiología al trabajo: “Bajo nivel de forma física en los adolescentes españoles. Implicaciones para la salud cardiovascular futura (Estudio AVENA). Rev Esp Cardiol 2005; 58: 898-909.

9. Primer premio al mejor trabajo de investigación deportiva. Instituto Andaluz del Deporte, ORDEN de 28 de julio de 2006. La fuerza joven de Andalucía en España y Europa.

191

Agradecimientos

La consecución del presente trabajo de Tesis Doctoral ha sido posible gracias a la participación de cientos de niños que corriendo, saltando y pedaleando han contribuido a mejorar mi: Capacidad, para poder entender las relaciones entre lo mecánico y lo biológico; Fuerza, para provocar al destino; Flexibilidad, para entender otras culturas, otras formas de vida, de trabajar; Velocidad de desplazamiento, sobre todo de los dedos al escribir en el ordenador; Coordinación,de mi vida, para tener presencia en dos países, y estar activo en todos los frentes; y Equilibrio, para compaginar lo profesional con lo personal.

Gracias también a la inestimable ayuda científica y humana de todos aquellos que forman parte de los grupos de investigación de los proyectos AVENA, EYHS y HELENA, y en especial a:

Manuel Castillo, por enseñarme a hacer las cosas mucho más sencillas. Gracias también por compartir conmigo su sentido práctico de la fisiología, y de la vida. Efectivamente, la recompensa de hacer el trabajo bien hecho es la oportunidad de hacer más trabajo bien hecho.

Ángel Gutiérrez, por sus lecciones, cada minuto que hemos pasado juntos, y por activar mi interés por la fisiología del ejercicio. Gracias también por animar siempre al duende “inquietud”.

Marcela González-Gross, por estar siempre “virtualmente” presente, y por darle el toque “femenino” (y alemán) a mi trabajo. Gracias también por guiarme en el camino de la ciencia, desde mi nacimiento.

Michael Sjöström, por darme la posibilidad de trabajar en su grupo. Gracias también por ponerme en órbita, y por enseñarme que la constancia y la ambición no están reñidas con la edad. Que pena que el día sólo tenga 24 horas. “Fit for fight?”

Olle Carlson, por las eternas discusiones estadísticas. Ciertamente, el lector nunca podrá adivinar cuanto tiempo tardamos en escribir el artículo, pero siempre podrá valorar la calidad del mismo.

Luis Moreno, por creer siempre en mi trabajo, en la labor de los más jóvenes. Gracias también por sus lecciones de coherencia, lógica y diplomacia.

Todos mis compañeros del grupo EFFECTS-262 y PrevNut, y en especial a Fran Ortega, José Luis Mesa, Ricardo Sola y Anita Hurtig-Wennlöf. Emma, gracias [thanks] por tu inestimable ayuda con el inglés!

Todos ellos también me enseñaron que sólo hay una forma de hacer las cosas…

Documents

La Condición Física como Determinante de Salud en Personas ... · Facultad de Medicina Universidad de Granada La Condición Física como Determinante de Salud en Personas Jóvenes